ESTUDO DA EFICIÊNCIAD O CHECKPOINTMITÓTICO NUMA …€¦ · Chromosome spread analysis also...
70
I MU p Zûoî ESTUDO DA EFICIÊNCIA DO CHECKPOINT MITÓTICO NUMA LINHA CELULAR DE MEDULOBLASTOMA MARIA JOANA ALMEIDA RODRIGUES BARBOSA Dissertação de Mestrado em Bioquímica QH605.2 BARm E 2009 Universidade do Porto Faculdade de Ciências uto de Ciências Biomédicas Abel Salazar 2009
Transcript of ESTUDO DA EFICIÊNCIAD O CHECKPOINTMITÓTICO NUMA …€¦ · Chromosome spread analysis also...
I MU p
Z û o î
ESTUDO DA EFICIÊNCIA DO CHECKPOINT MITÓTICO
NUMA LINHA CELULAR DE MEDULOBLASTOMA
MARIA JOANA ALMEIDA RODRIGUES BARBOSA
Dissertação de Mestrado em Bioquímica
QH605.2 BARm E
2009
Universidade do Porto
Faculdade de Ciências
uto de Ciências Biomédicas Abel Salazar
2009
ESTUDO DA EFICIÊNCIA DO CHECKPOINT MITÓTICO
NUMA LINHA CELULAR DE MEDULOBLASTOMA
MARIA JOANA ALMEIDA RODRIGUES BARBOSA
Dissertação de Mestrado em Bioquímica
^IVEHIOO' : S couro BIBLIOTECA Sala Co\oc.-LJ±±ï' ■
Depart. Química
Universidade do Porto
Faculdade de Ciências
Instituto de Ciências Biomédicas Abel Salazar
2009
MARIA JOANA ALMEIDA RODRIGUES BARBOSA
ESTUDO DA EFICIÊNCIA DO CHECKPOINT MITÓTICO NUMA
LINHA CELULAR DE MEDULOBLASTOMA
Dissertação de Candidatura ao grau de Mestre em Bioquímica da Universidade do Porto
Instituição de Acolhimento - Centro de Investigação em Ciências da Saúde (CICS), Instituto Superior de Ciências da Saúde - Norte (ISCS-N)/CESPU
Orientador - Doutor Hassan Bousbaa Categoria - Professor Associado
Afiliação - Departamento de Ciências e CICS, ISCS-N/CESPU
Co-orientador - Doutora Roxana Moreira Categoria - Professora Auxiliar Afiliação - Departamento de Ciências e CICS, ISCS-N/CESPU
2009
Acknowledgements
Ao Prof. Doutor Hassan Bousbaa e à Prof.a Doutora Roxana Moreira, por me
terem aceite como sua mestranda, pela disponibilidade incondicional ao longo de todo o
meu trabalho de mestrado, pela paciente revisão da presente tese e pelo exemplo dos
seus percursos científicos, profissionais e pessoais.
À Prof.a Doutora Odília Queirós, pelo acompanhamento próximo do curso dos
meus trabalhos e pelas valiosas sugestões para a sua progressão.
À Dr.a Olga Martinho e ao Prof. Doutor Rui Reis (ICVS, Universidade do Minho),
pela cedência das linhas celulares Daoy e S462 e do extracto de RNA total de astrócitos.
À Juliana e à Ana Vanessa, com quem tenho a sorte de partilhar um percurso
comum desde há muito, pela sua presença amiga, atenta e constante, pela sua boa-
disposição, tolerância, solidariedade e gratuidade, bem como pelos seus exemplos de
iniciativa e capacidade de trabalho.
À Kelly, pelo seu espírito activo, pela sua generosidade, boa-disposição e
amizade, e pela forma como ilustra o conceito de trabalho em equipa.
Ao Daniel, à Nita, à Vânia, à Catarina e à Ana Rita, pela amizade, incentivo e
afinidades que transcendem os tempos de faculdade.
À Rita e à Virgínia, pelos momentos calorosos passados dentro e fora do
laboratório, que com a sua amizade e partilha de experiências e conhecimentos ilustram
as potencialidades de uma equipa multidisciplinar.
À Vanessa, pelo companheirismo de uma vida e por tão bem saber ajudar-me a
relativizar os obstáculos que se me vão deparando. Pelo genuíno e gratificante exemplo
de vocação, gosto e dedicação pelo que se faz e por quem se faz.
Aos meus irmãos. À Marta, for always leading my way. Ao Eduardo, por ter
sempre algo a ensinar-me e com que me fazer rir. A ambos agradeço o apoio
incondicional, a partilha de interesses e o sentido de humor que tanto aprecio.
Aos meus pais, a quem devo a conquista de mais esta etapa, pela confiança que
sempre depositaram em mim, pelas palavras de incentivo e carinho que sempre
souberam dizer-me e pelo insubstituível exemplo de honestidade, persistência e trabalho.
Index
ACKNOWLEDGEMENTS 2
INDEX 3
ABSTRACT 4
RESUMO 5
ABBREVIATIONS 6
1. INTRODUCTION 8
1.1 The cell cycle 8
1.2 Kinetochore and microtubule structure and function 13
1.3 Cell cycle checkpoints 15
1.4 Checkpoint defects, aneuploidy and cancer 22
1.5 Medulloblastoma and Malignant Peripheral Nerve Sheath Tumour 25
1.6 Aim of the study 28
2. MATERIALS AND METHODS 29
2.1 Cell lines 29
2.2 Cell synchronization 30
2.3 Mitotic index determination 31
2.4 Immunofluorescence 31
2.5 Chromosome spread 32
2.6 Preparation of protein extracts 33
2.7 Western blotting 33
2.8 Real-Time PCR (two-step procedure) 35
2.9 Statistical analysis 39
3. RESULTS AND DISCUSSION 40
3.1 The mitotic checkpoint is defective in Daoy cell lines 40
3.2 Mitotic checkpoint proteins exhibit a normal subcellular distribution in the Daoy cell line 48
3.3 Daoy cell line shows altered expression of mitotic checkpoint genes 51
4. CONCLUSIONS 57
5. FUTURE PROSPECTS 59
6. REFERENCES 60
Abstract
Genetic instability constitutes the main cause of cancer development in man, with
chromosome instability (CIN) being one of its forms and deriving from abnormalities in
chromosome segregation during mitosis. In order to prevent CIN, cells possess a
mechanism of control, the mitotic checkpoint, which consists of a signal transduction
pathway activated to detect and repair eventual errors in chromosome segregation by
delaying anaphase onset until all chromosomes are appropriately attached in a bipolar
fashion to the mitotic spindle.
Medulloblastoma is the most frequent malignant central nervous system neoplasm
of childhood. Most paediatric medulloblastomas have abnormal karyotypes, including
aneuploidy. The present work aimed at detecting possible defects in mitotic checkpoint
activity and the underlying molecular alterations in the Daoy medulloblastoma cell line.
HeLa (cervical cancer) and S462 (Malignant Peripheral Nerve Sheath Tumour) cell lines
were used as controls for the mitotic checkpoint efficiency.
Mitotic checkpoint competence was assessed through incubation with microtubule-
disrupting nocodazole. Upon exposure to the drug, both HeLa and S462 cells accumulate
in mitosis and eventually die as a response to microtubule depolymerisation, reflecting the
efficiency of their mitotic checkpoint. Daoy cells, in turn, are capable of exiting mitosis and
escaping from cell death in the presence of nocodazole, being able to survive even after
prolonged incubation with the drug, an ability that is supported by their comparatively
lower mitotic index. Chromosome spread analysis also reveals significant sister chromatid
separation upon incubation with nocodazole, confirming that Daoy cells prematurely exit
mitosis. Immunofluorescence analysis has shown that subcellular localisation of mitotic
checkpoint proteins has no detectable changes in Daoy cells. However, Western blotting
and quantitative Real-Time PCR assays denote alterations in their expression levels. In
particular, BubR1, Bub3 and Mad2 are underexpressed, while there is Cdc20 and Chfr
overexpression.
Alterations in these proteins' expression account for the observed mitotic
checkpoint defects in Daoy cells and may act as aneuploidy promoters and participate in
medulloblastoma's progression and/or malignancy.
Resumo A instabilidade genética constitui a principal causa de desenvolvimento de cancro
no Homem, sendo a instabilidade cromossómica, resultante de anomalias na segregação
cromossómica ocorrida aquando da mitose, uma das suas formas. No sentido de
evitarem a instabilidade cromossómica, as células dispõem de um mecanismo de
controlo, o checkpoint mitótico, que consiste numa via de transdução de sinal activada
para detectar e reparar eventuais erros na segregação cromossómica. Esta via actua
atrasando o início da anafase até que todos os cromossomas se encontrem ligados de
forma adequada e bipolar ao fuso mitótico.
Os meduloblastomas constituem as neoplasias malignas do sistema nervoso
central mais frequentes em crianças. A maioria dos meduloblastomas pediátricos revela
anormalidades cariotípicas, incluindo aneuploidia. O presente trabalho teve como
objectivos a detecção de possíveis falhas na actividade do checkpoint mitótico e as
alterações moleculares subjacentes na linha celular Daoy, derivada de meduloblastoma.
As linhas celulares HeLa (cancro cervical) e S462 (Malignant Peripheral Nerve Sheath
Tumour) foram utilizadas como controlos da eficiência do checkpoint mitótico.
A actividade do checkpoint mitótico foi avaliada através da incubação com
nocodazole, uma droga despolimerizadora de microtúbulos. Após exposição à mesma, as
células HeLa e S462 acumulam em mitose e eventualmente morrem como resposta à
despolimerização dos microtúbulos, reflectindo a eficiência do seu checkpoint mitótico.
Por sua vez, as células Daoy mostraram-se capazes de sair da mitose e de escapar à
morte celular na presença de nocodazole, sobrevivendo mesmo após incubação
prolongada, o que foi também corroborado pelo seu índice mitótico comparativamente
menor. Ensaios de chromosome spread revelaram igualmente uma significativa
percentagem de células com separação de cromátidas irmãs mediante incubação com
nocodazole, confirmando que as células Daoy saem prematuramente da mitose. A
análise por imunofluorescência revelou a ausência de alterações detectáveis na
distribuição subcelular de proteínas do checkpoint mitótico. Contudo, ensaios de Western
blotting e de Real-Time PCR quantitativo evidenciaram alterações nos seus níveis de
expressão, demonstrando, mais concretamente, a sob-expressão de BubR1, Bub3 e
Mad2, assim como a sobre-expressão de Cdc20 e de Chfr.
As alterações na expressão de proteínas do checkpoint mitótico justificam as
falhas na sua actividade nas células Daoy, podendo constituir um mecanismo promotor
de aneuploidia e contribuir para a progressão e/ou malignidade dos meduloblastomas.
- 5 -
Abbreviations
APC: Adenomatous Polyposis Coli
APC/C: Anaphase-Promoting
Complex/Cyclosome
bp: base pairs
BRCA1 : Breast Cancer 1
Bub: Budding uninhibited by benzimidazole
BubR1: Bub1-related protein
Cdc20: Cell-division-cycle 20 homologue
CDK: Cyclin-Dependent Kinase
cDNA: complementary Deoxyribonucleic
Acid
CENP-E: Centromere Protein E
Chfr: Checkpoint with fork-head associated
and ring finger
CIN: Chromosomal Instability
C02: Carbon Dioxide
DAPI: 4', 6-diamidino-2-phenylindole
D-MEM: Dulbecco's Modified Eagle
Medium
DMSO: Dimethylsulfoxide
DNA: Deoxyribonucleic Acid
DTT: Dithiothreitol
ECL: Enhanced Chemiluminescence
EDTA: Etylenediaminetetracetic Acid
FBS: Foetal Bovine Serum
GAPDH: Glyceraldehyde-3-phosphate-
dehydrogenase
HRP: Horseradish peroxidase
H202: Hydrogen Peroxide
Kbp: Kilo base pairs
kDa: kiloDaltons
M: molar (mol dm"3)
Mad: Mitotic arrest deficient
MAPK: Mitogen-Activated Protein Kinase
MCC: Mitotic Checkpoint Complex
u.g: micrograms
[il: microlitres
pm: micrometres
pM: micromolar
ml: millilitres
mm: millimetres
mM: millimolar
Ml: Mitotic Index
MPNST: Malignant Peripheral Nerve
Sheath Tumour
Mps1 : Monopolar spindle 1
MTOC: Microtubule-organizing centre
nm: nanometres
nM: nanomolar
OD: Optical Density
PBS: Phosphate-Buffered Saline
PBST: Phosphate-Buffered Saline and
Tween-20
PCR: Polymerase Chain Reaction
PFA: Paraformaldehyde
Plk1: Polo-like kinase 1
PMSF: Phenylmethylsulfonyl Fluoride
PTEN: Phosphatase and Tensin
Homologue
RB1 : Retinoblastoma 1
RNA: Ribonucleic Acid
RNAi: RNA interference
RNase: Ribonuclease
RZZ: ROD-ZW10-ZWILCH complex
rpm: rotations per minute
SAC: Spindle Assembly Checkpoint
SCS: Sister Chromatid Separation
SD: Standard Deviation
SDS: Sodium Dodecyl Sulfate
SDS-PAGE: Sodium Dodecyl Sulfate-
Polyacrylamide Gel Electrophoresis
Shh-Ptch: Sonic hedgehog-Patched
pathway
TBS: Tris-Buffered Saline
TBST: Tris-Buffered Saline and Tween-20
TP53: Tumour Protein p53
Tris: Tris-(hydroxymethyl)aminemethane
ZW10: Zeste White 10
- 7 -
1. INTRODUCTION
1.1 The cell cycle
The cell cycle is a ubiquitous and complex process of major importance in the
growth and proliferation of cells, development of all living organisms, regulation of DNA
damage repair and tissue regeneration (Schafer, 1998). It is a sequential phenomenon,
orchestrated and controlled by numerous regulatory proteins that direct the cell through a
specific series of highly coordinated events in a generally invariant order (Schafer, 1998;
Clarke and Giménez-Abián, 2000). Nonetheless, cell cycle-related deregulations have
been shown to be associated with a number of pathologies, particularly with cancer
(Garrett, 2001; Vermeulen era/., 2003; Kops et al., 2005; Bharadwaj and Yu, 2004). In
particular, aneuploidy - an abnormal chromosomal content - has been observed in a
variety of tumour cells, suggesting that defects in the mechanisms that control cell division
and chromosome segregation may be involved in tumorigenesis.
The somatic cell cycle may be defined as the period comprised between two
mitotic divisions. It may be divided into two main stages: the M phase, which refers to the
period where division actually occurs, and the interphase, which corresponds to the time
from the end of one mitotic division until the start of the next. In order to be able to divide
and originate two new identical daughter cells, each of them bearing a diploid set of
chromosomes, a eukaryotic somatic cell must double its size and its contents. In reality,
the doubling of its size is a continuous process, deriving from constant transcription and
translation of genes encoding for the proteins that are needed to a specific phenotype.
However, DNA replication is confined to a particular phase of the cycle. Interphase
encompasses the G1, S and G2 periods, which were actually identified accordingly with
the timing of DNA synthesis. In G1, there is RNA and protein synthesis, but no DNA
replication, with no apparent changes in cell morphology. The S phase, in turn, lasts from
the beginning of DNA replication until it is all replicated. In this step, the total DNA content
increases from the diploid value 2n to the fully replicated value of An and the cell
increases noticeably in size, in part because of proteins that accumulate to match DNA
synthesis. G2 comprises the period between the S phase and mitosis, in which the cell
possesses two complete sets of chromosomes. Thus, G1 and G2 stand for the 'gap'
periods in which there is no DNA replication, whilst the S phase refers to the synthetic
- 8 -
period when DNA is replicated. Virtually, all synthetic activities are suspended during
mitosis (Vermeulen era/., 2003; Lewin, 2004).
It must be stated that, when in G1, cells may alternatively withdraw from the cycle
into the GO phase, a quiescence state, for an indefinite period of time (e.g., when nutrient
supply is not favourable); conversely, they may reactivate and re-enter the cell cycle in G1
phase at any time (fig. 1.1).
Daughter / cells
G, Chromosome decondensation, re-formation of
Chromosome ""clear envelope, condensation, cytokinesis nuclear envelope breakdown, chromosome segregation
Sister chromatids
Fig. 1.1: The eukaryotic cell cycle.
Interphase comprises G1, S and G2
phases. In G1 (6-12 hours) and in G2
(3-4 hours), there is RNA and protein
synthesis; in S phase (6-8 hours), there
is also DNA replication. One cell cycle
is separated from the next by mitosis,
the M phase (1 hour), in which
karyocinesis (nuclear division) and
cytokinesis (cell division) yield two
identical daughter cells. Terminally
differentiated cells may alternatively
enter a quiescent phase (GO) for an
indefinite period of time (Lodish et al.,
2000).
DNA synthesis
Successful division is dependent on the coordinated duplication and segregation of
all cell components. This, in turn, is under the control of a family of highly regulated
protein kinases known as cyclin-dependent kinases (Cdks). As their name implies, their
activity requires the binding to a cyclin subunit (McGowan, 2003; Murray, 2004). Cyclins
were originally characterized as proteins whose levels changed periodically during the cell
cycle; in fact, Cdk most fundamental level of control is cyclin intermittent expression (fig.
1.2) (Schafer, 1998; McGowan, 2003; Murray, 2004; Sullivan and Morgan, 2007).
The first cyclin to be expressed following the quiescence state is cyclin D, which
assembles with Cdk4 and Cdk6 (fig. 1.2). This complex enters the nucleus and promotes
phosphorylation of the retinoblastoma protein (Rb) and the related pocket proteins p107
and p130. Rb phosphorylation stimulates the activation of the E2F family of transcription
factors and induces the transcription of fundamental proteins for G1 and S phase
(Schafer, 1998; McGowan, 2003). The same mitogenic stimuli that promote cyclin D
production also promote cyclin E expression, as well as of two cyclin-dependent kinase
inhibitors, p21cip1 and p27kip1 (Sherr and Roberts, 1999; Egozi et ai, 2007). These latter
proteins are necessary for the efficient assembly and nuclear import of active cyclin D-
Cdk4 complexes and delay activation of cyclin E-Cdk2 in early G1 (McGowan, 2003).
- 9 -
The activation of cyclin E-Cdk2 is concomitant with the beginning of DNA
replication. Cyclin E-Cdk2 phosphorylates the inhibitor protein p27, facilitating its
ubiquitination and subsequent degradation (Sherr and Roberts, 1999; Egozi et al., 2007).
Cyclin E-Cdk2 also autophosphorylates on cyclin E, similarly targeting it to degradation.
For this reason, the kinase complex is said to be both self-activating and self-limiting
(Schafer, 1998; McGowan, 2003).
Initiation and completion of DNA replication is also dependent on another complex,
cyclin A-Cdk2 (Cuomo et a/., 2008). Cyclin A expression follows that of cyclin E, at the
G1-S boundary. Both cyclin A-Cdk2 and cyclin E-Cdk2 are crucial to assure that DNA
replication occurs only once per cycle. Moreover, cyclin A-Cdk2 increases transcription of
histones and other genes, ensuring the efficient accomplishment of S phase by furnishing
the material required for replication (McGowan, 2003).
Mitotic checkpoint
G2 DNA damage checkpoint
CDK1
CDK4/COK6
CycNnD1,D2, D3
Restriction point
CDK2 Cyclin El, E2 G1/S DNA damage
checkpoint
CDK2 Cyclin AI, A2
S Phase DNA damage checkpoint
Fig. 1.2: Cell cycle regulat ion by Cdk-cycl in complexes. The approximate time of activity of the different
Cdk-cyclin complexes throughout the cell cycle, based on studies with mammalian cells, is depicted. Shapes
outside the cycle reflect increases and decreases of the corresponding Cdk-cyclin activities. Cell cycle
checkpoints are also represented (adapted from Garrett, 2001; van den Heuvel, 2005).
The profound morphological changes that accompany mitosis are under the
regulation of Cdk1(Cdc2) in association with cyclin A and cyclin B. Expression of cyclin B
lags behind that of cyclin A, peaking late in S phase and remaining high during G2 and
mitosis. Cyclin B-Cdk1 is largely maintained in a phosphorylated inactive state, and its
posterior dephosphorylation and activation coincides with the morphological alterations
that are typical of mitosis (McGowan, 2003). It is known that Cdk1, in association with a
second cyclin B subunit - cyclin B2 - , plays a role in the division of single-copy organelles
such as the endoplasmic reticulum, making sure that cell division occurs in a way that
- 1 0 -
supplies sufficient amounts of these organelles to each daughter cell, so that they can be
rebuilt. Multi-copy organelles (e.g., mitochondria) are homogeneously distributed in the
cytoplasm so that each daughter cell can inherit roughly the same amount of them
(McGowan, 2003).
When cell cycle proteins have entirely performed their function, they must be
degraded so that the events they elicited do not reoccur improperly. Ubiquitin-mediated
degradation is the main pathway by which they are eliminated (Murray, 2004). Cyclins are
targeted for ubiquitination by their PEST sequences, protein motifs rich in proline (P),
glutamate (E), serine (S) and threonine (T) (Schafer, 1998). Protein tagging with ubiquitin
residues signalises it to degradation by the 26S proteasome, a multisubunit cellular
complex that is specialised in protein unfolding and proteolysis (Schafer, 1998).
1.1.1 Mitosis Following interphase, cells enter mitosis, which, though a short-termed phase
when considering the temporal context of the whole cell cycle, is remarkably important for
cell division and normal proliferation since it ensures accurate segregation of genetic
material into the two daughter nuclei. It is traditionally subdivided into five stages:
prophase, prometaphase, metaphase, anaphase and telophase (fig. 1.3) (Nigg, 2001;
McGowan, 2003; Lewin, 2004; Kops etal., 2005).
At prophase, chromatin condensates, originating recognizable chromosomes. Until
this phase, it was a compact mass with no visible change in its condensation state.
Previously duplicated centrosomes separate and define the future poles of the mitotic
spindle (Compton, 2000; Nigg, 2001; McGowan, 2003). In turn, numerous highly dynamic
microtubules nucleate from centrosomes (Lewin, 2004; Doxsey et al., 2005; Kops et a/.,
2005). This is the reason why centrosomes are also known as microtubule organizing
centres (MTOCs) (Crasta etal., 2008; Kwon etal., 2008).
At prometaphase, the nuclear envelope breaks down and the consequently
dispersed chromosomes attach, through their kinetochores, to microtubules of the mitotic
spindle through a 'search and capture' process (Gadde and Heald, 2004; Odde, 2005;
Pinsky and Biggins, 2005; Yang er a/., 2008). The Golgi apparatus and the endoplasmic
reticulum also break down into small membrane vesicles in this stage (Nigg, 2001;
McGowan, 2003; Lewin, 2004; Kops etal., 2005).
-11 -
Sister chromatids
Fig. 1.3: The five stages of mitosis. Mitosis comprises orderly sequenced events. Nuclear envelope
breakdown, chromosome condensation and centrosome separation take place at prophase, while in
prometaphase chromosomes are captured by microtubules growing from the separated centrosomes and start
to congress. At metaphase, all chromosomes are aligned at the equatorial plate. Upon chromatid separation
and spindle pole elongation, occurring at anaphase, two daughter cells are originated at telophase, prior to
cytokinesis. Then, cells return to interphase (Cheeseman and Desai, 2008).
At metaphase, all chromosomes have undergone proper bipolar attachment and
alignment in the metaphase plate. Sister chromatid cohesion is then suddenly lost
(section 1.3.1.1) and anaphase is initiated, comprising spindle elongation and
consequent separation of sister chromatids. In the end of anaphase, the plasma
membrane starts to evidence a certain degree of invagination around the spindle
midzone, the contractile ring. Some authors distinguish between anaphase A and
anaphase B, considering that the former is characterized by chromosome migration
towards the poles and that, in the latter, further separation of the poles towards the cell
cortex occurs (Nigg, 2001). Telophase is characterized by the progression of the
contractile ring, which is composed of actomyosin fibers that extend around the equator of
the dividing cell and then pinch inward until they contact a group of microtubules that run
between the poles, giving rise to a structure connecting the future two daughter cells - the
midbody. This structure is dissociated, ultimately leading to cytoplasm partitioning and to
the formation of two new daughter cells, in a process called cytokinesis. In the meanwhile,
chromatin has decondensed and the nuclear envelope is completely reconstituted around
-12 -
the two daughter groups of chromosomes (Nigg, 2001; McGowan, 2003; Lewin, 2004;
Kopsefa/., 2005).
1.2 Kinetochore and microtubule structure and function
Genetic integrity maintenance is ensured by proper chromosome segregation
during mitosis, which takes place on a bipolar spindle-shaped structure composed of
microtubules. These interact with chromosomes through specialised complexes, the
kinetochores, which are located in the centromeres, the sites of primary constriction of
condensed chromosomes (Chan et al, 2005; Tanaka and Desai, 2007; Cheeseman and
Desai, 2008).
In effect, kinetochores constitute the macromolecular structures assembled on
opposite sides of the centromere that form the interface between the chromosomes and
the microtubules of the mitotic spindle, therefore playing a crucial role during mitosis
(Chan et al., 2005; Cheeseman and Desai, 2008). Kinetochores comprise several highly
conserved subcomplexes whose functions include attachment of chromosomes to the
spindle microtubules, monitoring these attachments - and, if defects are detected,
initiating a signalling checkpoint pathway to prevent cell cycle progression - , kinetochore
assembly and specification (that is, determining where on the chromosome will the
kinetochore be assembled), and participating in chromosome movement along the spindle
(Cheeseman and Desai, 2008).
Microtubules, in turn, are 25-nm diameter hollow polymers, comprised of 12-15
protofilaments that are formed by head-to-tail association of a/B-tubulin dimers. They
alternate between phases of growth and shrinkage, a phenomenon referred to as
'dynamic instability'. Because tubulin dimers have a fixed orientation, the microtubule
lattice is polar, influencing movements along the spindle (Maiato etal, 2004; Cheeseman
and Desai, 2008).
The structure of animal kinetochores may be described as a trilaminar stack of
plates (fig. 1.4). The inner kinetochore is composed of repetitive DNA sequences and
assembles into a specialized highly folded form of heterochromatin, containing
nucleosomes with at least one specialised histone and auxiliary proteins, which persists
all cell cycle. This kinetochore-adjacent single inner plate is thus a very ordered, modular
structure (Maiato et al, 2004). On the surface of the chromosome, there is a
proteinaceous outer plate whose dynamic components assemble and function only during
mitosis, more specifically at about the time of nuclear envelope breakdown (Maiato et al,
2004). This outer plate displays about 15 to 20 end-on attachment sites for the plus ends
- 1 3 -
of microtubules (called kinetochore microtubules, kMTs) and is 50-60 nm-thick
(Cheeseman and Desai, 2008). The outermost regions of the kinetochore constitute a
fibrous corona, which is composed of several resident and transient components that are
responsible for microtubule attachment and behaviour and for spindle checkpoint activity
(Maiato et al., 2004; Chan et al., 2005). In cultured mammalian cells, duplicated
kinetochores belonging to sister chromatids separate from one another during mid-late
G2. Upon nuclear envelope breakdown, they acquire a mature laminar structure, in a
process that depends on inner plate elements (Maiato et al., 2004). The central
kinetochore is the region comprised between inner and outer plates; it seems to be less
dense than its neighbours and its protein composition has not been determined yet (Chan
etal., 2005; Cheeseman and Desai, 2008).
As far as kinetochore protein composition and interactions are concerned, it should
be stated that they constitute a very broad subject that is continuously being updated. In
humans, about 80 kinetochore proteins are known today (Cheeseman and Desai, 2008).
Nevertheless, many components remain to be unravelled and the interdependencies
established so far may suffer from slight changes. Fig. 1.4 depicts some of the most
important kinetochore protein components and their respective functions in the cell cycle
(see Maiato etal., 2004 for a review).
Fig. 1.4: Animal kinetochore
organization. Some of
kinetochore protein
constituents, as well as their
localisation and main
functions are depicted.
MT: microtubules (Maiato et
ai, 2004).
< > < > INNER OUTER
KINETOCHORE KINETOCHORE
Upon nuclear envelope breakdown, at prophase-to-prometaphase transition,
usually microtubules become mono-oriented by one (monotelic attachment) or both
(syntelic attachment) sister kinetochores. However, chromosome bi-orientation is required
for accurate chromosome segregation, implying that one sister kinetochore must be
bound to microtubules irradiating from solely one pole, whilst the other must be attached
CENTROMERE KINETOCHORE
■* m
m PLUS ENDS FIBROUS CORONA
OUTER PLATE
CENP-B MCAK INCENP Aurora B Survivin Borealin'Dasra B ICIS
Chromatid pairing, structural support, & MT attachment error correction
INNER PLATE INTERZONE
CENP-A CENP-C CENP-G CENP-H CENP-l/hMis6 CeKNL-1 hMis12
3F3/2 ant lgsm CENP-A CENP-C CENP-G CENP-H CENP-l/hMis6 CeKNL-1 hMis12
Tension receptors & checkpoint signalling
CENP-A CENP-C CENP-G CENP-H CENP-l/hMis6 CeKNL-1 hMis12
Kinetochore assembly &size determination
NdcSO/HacI RanQAPI Nuf2 RanBP2 Spc24 CENP-F Spc25 Plk BUB1 PP1 BUBR1 Zw10 BUB3 Zwint-1 MPS1 Rod MAD1 Dynoin MAD2 LIS1 Cdc20 CLASPS CENP-E CLIP-170
EBI APC
MT attachment, regulation of MT dynamics. & checkpoint signalling
- 1 4 -
to microtubules solely from the opposite pole. In other words, an amphitelic attachment
must be achieved. Merotelic attachments, in turn, occur when one or both of the sister
kinetochores are attached to microtubules from opposite poles (fig. 1.5) (Hauf and
Watanabe, 2004; Maiato et al., 2004; Musacchio and Salmon, 2007). Because it still
originates tension, this situation is not detected by the spindle assembly checkpoint;
instead, it is corrected by Aurora B/lpl1 (Musacchio and Salmon, 2007).
Orientation Sensed by checkpoint
Mis-segregation error if uncorrected
Amphitelic
Monotelic
Syntelic (ERROR)
Merotelic (ERROR)
NO
YES
YES
NO
None Spindle pole
O
o
o Hi* Kinetochore checkpoint activity f High Q Medium Q Oft
Fig. 1.5: Schematic representat ion of the most common k inetochore attachment errors. The
corresponding extent of checkpoint activation and their consequences if they are not repaired before
anaphase onset are represented. In an amphitelic attachment, sister kinetochores are attached to opposite
poles. In a monotelic attachment, only one of the sister kinetochores is attached to one pole, while in a
syntelic attachment both sister kinetochores are attached to the same pole. Finally, if there is one sister
kinetochore attached to both poles, the attachment is considered to be merotelic. In monotelic and syntelic
attachments, sister kinetochores are mono-oriented, while in amphitelic and in merotelic attachments, they are
said to be bi-oriented (Maiato etal., 2004).
1.3 Cell cycle checkpoints
The orderly progress through the cell cycle is assured by specific 'checkpoints' -
control loops that make the initiation of each next event dependent on the proper
accomplishment of the previous, being commonly defined as mechanisms that establish a
dependence relationship between two cellular processes that are biochemically unrelated
(Schafer, 1998; Clarke and Giménez-Abián, 2000). In other words, these checkpoints
confirm cells' ability and readiness to proceed to a new stage, being superimposed on the
cell cycle and acting directly on the regulatory molecules that control progression through
it (Lewin, 2004). A checkpoint is made up of three components that act sequentially. The
first component to operate is a sensor mechanism that detects incorrect or incomplete cell
- 1 5 -
cycle events, triggering a signalling pathway, the second component, which may
culminate in the activation of an effector (the third and last component). This ultimate
target promotes cell cycle arrest until the error is solved (Garrett, 2001).
In effect, four control points must be highlighted: DNA damage checkpoints
occurring at G1/S, S and G2/M and the mitotic checkpoint (fig. 1.2). DNA damage is
checked at every stage of the cycle, but especially at G1/S and G2/M transitions. At S
phase, the cell verifies DNA integrity and the completion of its replication, which is a
prerequisite for the division of probably all somatic eukaryotic cells. The G2/M checkpoint
assures that every DNA sequence has been replicated totally and only once and that the
cell does not begin mitosis until replication is completed (Kaufmann and Paules, 1996;
Garrett, 2001).
The spindle checkpoint ensures the assembly of a functional mitotic spindle and
inspects kinetochore attachments to microtubules, which is mentioned in greater detail in
section 1.3.1.1 (Garrett, 2001; Lewin, 2004). In case of functional DNA damage and/or
spindle assembly checkpoints, the cell is able to correct the errors or, if they are too
extensive or irremediable, it may undergo apoptosis, senescence, or mitotic catastrophe,
for instance (Schmit and Ahmad, 2007).
Chfr (Checkpoint with Forkhead-associated and RING finger domains), reported
as an early mitotic checkpoint protein, is an E3 ubiquitin ligase responsible for targeting
proteins such as Plk1 and Aurora A for degradation (Kang et al., 2002; Privette and Petty,
2008). In case of mitotic stress (for instance, in case of drug-induced microtubule
depolymerisation), it has been shown to act in late G2, to delay prophase and entry into
metaphase by inhibiting chromosome condensation, nuclear envelope breakdown and by
excluding cyclin B1/Cdk1 from the nucleus (Scolnick and Halazonetis, 2000; Chaturvedi et
al., 2002; Gong, 2004; Loring étal., 2008; Privette and Petty, 2008). Other studies support
its role as a potent tumour suppressor and its importance, later in mitosis, for the control
of centrosome separation, mitotic spindle formation, chromosome segregation, and mitotic
spindle assembly checkpoint (Scolnick and Halazonetis, 2000; Privette et al., 2007;
Privette and Petty, 2008). In effect, Chfr was found to be required for proper BubR1 and
Mad2 localization to the kinetochores and for Mad2/Cdc20 complex formation (section
1.3.1) (Privette et al., 2007; Privette et al., 2008). Several cancer cell lines - namely from
lung, colon and esophageal neoplasms - display Chfr gene down-regulation or
methylation-induced inactivation (Corn et al., 2003; Toyota et al., 2003), which correlate
with tumour higher invasiveness and aneuploidy (Privette et a/., 2007; Loring et al., 2008;
Privette et al., 2008). Chfr is thus implied as a regulator of genomic stability (Privette and
Petty, 2008).
-16 -
1.3.1 The spindle assembly checkpoint The spindle assembly checkpoint (SAC) is a constitutive surveillance mechanism
in eukaryotic dividing cells that prevents chromosome mis-segregation by delaying the
metaphase-to-anaphase transition until all chromosomes are correctly connected to the
microtubule network, bi-oriented and aligned at the metaphase plate (Rieder et al., 1994;
May and Hardwick, 2006; Logarinho and Bousbaa, 2008). It represents the primary cell-
cycle control mechanism in mitosis, being activated immediately after mitosis or meiosis
entrance, every cell cycle (Kops et al., 2005). Regular functioning of this complex
signalling cascade is the main responsible for the even partition of the genetic material
into the two daughter cells and for the effective reduction of the error rate occurring during
cell division.
Ever since attention has been drawn to the role of SAC, a set of proteins have
been proven to be tightly associated with it and crucial to its functions (table 1.1). These
'core checkpoint proteins' comprise members of the Mad (Mitotic arrest deficient, Mad1-3)
and Bub (Budding uninhibited by benzimidazole, Bub1-3) families, which have been
initially identified by yeast mutagenesis screens for mutants unable to survive a temporary
exposure to microtubule toxins nocodazole and benzimidazole (Earnshaw and Mackay,
1994; Bharadwaj and Yu, 2004). Mps1 (monopolar spindle 1) protein was also shown to
play a relevant role in the spindle checkpoint, besides its initially identified function as a
requirement for the proper assembly of bipolar spindles (Wang and Burke, 1995;
Bharadwaj and Yu, 2004; Winey and Huneycutt, 2002). Subsequently, homologues for
these proteins have been identified in higher organisms (including mammals) (Bharadwaj
and Yu, 2004).
These proteins have been proven to share a high degree of both sequence and
functional homology with their yeast counterparts, as functional disruption studies through
dominant-negative mutants, antibody injection or RNA interference (RNAi) completely
compromised the spindle checkpoint activity, causing chromosome mis-segregation,
aneuploidy and mitotic arrest inability in the presence of microtubule poisons such as
nocodazole and taxol (Bharadwaj and Yu, 2004).
-17 -
Protein
Character ist ics Binding Partners
Comments Protein Molecular weight (kDa)
Other Binding Partners
Comments
Bub1 122 serine /
threonine kinase
Bub3
Inhibits Cdc20 by phosphorylation. Required for recruiting other checkpoint proteins differs
depending on system. Kinase activity is not required for checkpoint arrest.
BubR1 120 serine /
threonine kinase
CENP-E, Bub3, Cdc20
Part of APC/C inhibitory complex. Directly binds to Cdc20 and inhibits APC/C activity. Interacts
with Mad2 and Cdc20 to form the MCC. C-terminal kinase domain of BubR1 is activated by CENP-E. Yeast Mad3, the functional equivalent of BubR1, lacks the kinase domain, which is not
required for the checkpoint.
Bub3 37
structure determined:
7-bladed propeller of
WD40 repeats
Bub1, BubR1 Part of APC/C inhibitory complex. Localizes Bub1 and BubR1 to KTs.
Mad1 83 coiled coil Mad2
Directly recruits Mad2 to unattached KTs. Localizes Mad2 to the NP in interphase; function at the NP is unknown. Binds to Bub1 and Bub3
upon checkpoint activation in budding yeast.
Mad2 23 structure determined
Mad1,Cdc20, CIVT^/pSI00™'
Part of APC/C inhibitory complex. Directly binds to Cdc20 and inhibits APC/C activity. Occurs in two conformations ('closed' C-Mad2 on binding
Mad1 or Cdc20, or 'open' 0-Mad2 when unbound); interacts with Cdc20 and
BubR1/Mad3 to form the MCC.
Mps1 97 dual-specificity kinase Unknown
Phosphorylates Mad1 in vitro. Excess activates the checkpoint.
Required for recruitment of Mad1, Mad2 and CENP-E to the kinetochore. Required for spindle
pole body duplication.
CENP-E 312
plus-end directed
microtubule motor
BubR1
Binds to BubR1; stimulates BubR1 kinase activity. Required for capture and stabilization of microtubules at the kinetochore. Only found in
higher eukaryotes.
! ZW10 89 none identified ROD, Zwilch Part of complex that recruits the Mad1/Mad2 heterodimer to unattached KTs.
ROD 251 none identified ZW10, Zwilch Part of complex that recruits the Mad1/Mad2 heterodimer to unattached KTs.
Zwi lch 67 none identified ROD, ZW10 Part of complex that recruits the Mad1/Mad2 heterodimer to unattached KTs.
Table 1.1: Mitotic checkpoint core and related proteins/complexes. Most significant properties, interactions
and functions are emphasized. KTs: kinetochores; MCC: Mitotic Checkpoint Complex; NP: Nuclear Periphery
(adapted from Lew and Burke, 2003; Chan era/., 2005; Kops era/., 2005; May and Hardwick, 2006).
- 1 8 -
Other SAC proteins include the Aurora kinase B, the plus-end directed kinesin
motor protein CENP-E (Centromere protein E), MAPK (mitogen-activated protein kinase)
(Kops et al., 2005), as well as the minus-end directed motor dynein and the proteins with
which it interacts - dynactin, CLIP170 and LIS1, and the RZZ (ROD-ZW10-ZWILCH)
complex (Karess, 2005; Schmidt and Medema, 2006; Logarinho and Bousbaa, 2008). The
fact that these proteins do not have yeast orthologues suggests a more complex
regulation of mitosis in higher eukaryotes, in which they are actually required for normal
mitotic timing, as opposite to what happens in yeast (Taylor, 2004; May and Hardwick,
2006). A description of some proteins involved in SAC activity is provided on table 1.1.
In addition to the proteins that have been mentioned so far, recent works have
shed light on other components that seem to play a role in checkpoint signalling in
metazoans. That is the case of the microtubule-affinity regulating kinase (MARK)-family
kinase TA01, Polo-like kinase-1-interacting checkpoint helicase (PICH), and the
deubiquitylase USP44 (involved in the regulation of the cytosolic ubiquitination status of
Cdc20). Dynein at kinetochores has been suggested to function in checkpoint inactivation,
while a protein named Spindly was recently associated with dynein recruitment and
checkpoint silencing (Cheeseman and Desai, 2008).
1.3.1.1 The spindle assembly checkpoint acts by preventing anaphase onset
Sister chromatids are held together by a complex of cohesin proteins, formed in S
phase, which acts as a sort of 'glue' (through DNA cross-linking) (Marangos and Carroll,
2008). It is located at diverse sites along a pair of sister chromatids, appearing to be
centrally localised between the chromatids (Lewin, 2004). Anaphase onset implies
degradation of one of cohesin's subunits, Scc1, which is promoted by the proteolytic
activity of separase (Nasmyth, 2005). Separase, in turn, is normally kept inactive by
securin. When all chromosomes are correctly attached to microtubules, SAC requirements
are fulfilled and securin is then ubiquitinated by the anaphase promoting
complex/cyclosome (APC/C), a multi-subunit E3 ubiquitin ligase (Morgan, 1999; Reddy et
al., 2007; Stegmeier et al., 2007; Logarinho et al., 2008). Securin ubiquitination targets it
to degradation by the 26S proteasome. Separase is then activated and Scc1, an
important component of the cohesin complex, is cleaved, no longer holding the
chromatids together. Thus, they become free to segregate on the spindle, initiating
anaphase (fig. 1.6) (Morgan, 1999; Bharadwaj and Yu, 2004; Lewin, 2004; Nasmyth,
2005). Degradation of cyclin B1, another important mitotic protein, is also accomplished
through APC/C-mediated ubiquitination, leading to the inactivation of cyclin-dependent
- 1 9 -
kinase 1 (Cdk1). Cyclin B1 destruction makes it possible to reverse all the
phosphorylations that elicited mitosis, thus propitiating mitotic exit (Lewin, 2004; May and
Hardwick, 2006; Schmidt and Medema, 2006; Logarinho and Bousbaa, 2008).
Securin ubiquitination by APC/C is dependent on the binding of the WD40-repeat-
containing cell-division-cycle 20 homologue (Cdc20) protein, which is responsible for
substrate recruitment to APC/C. In fact, the complex between APC/C and Cdc20
(APC/CCdc20) is the most downstream target of the spindle assembly checkpoint
(Bharadwaj and Yu, 2004). Cdc20 is actually necessary for the APC to degrade securin,
which, as already mentioned, controls metaphase-to-anaphase transition (Lewin, 2004).
Ubiquitin ligase activity is reinforced by the phosphorylation of Cdc20 and APC core
components by cyclin B-Cdk1 (McGowan, 2003). In turn, Mad2, the protein that binds and
sequesters Cdc20, is critical for SAC functioning. It can adopt either an open conformation
(0-Mad2) or a closed form (C-Mad2). The latter is a more potent inhibitor of APC/C in vitro
given its higher affinity for Cdc20, and Mad2 binding to Mad1 is thought to promote
conformational switch to the closed form (fig. 1.6) (Chan et al., 2005; Musacchio and
Salmon, 2008). In effect, it was proven that Mad2 binding to Mad1 in a 2:2 tetramer is
required for in vivo Mad2:Cdc20 complexation (Mapelli et al., 2006).
Cdh1, like Cdc20, has also been reported as an APC/C regulatory protein,
interacting and presenting certain substrates to APC/C for ubiquitination (May and
Hardwick, 2006; Crasta et al., 2008). It is also required for the APC/C to degrade cyclin B,
therefore being a prerequisite for the cell to exit mitosis (Lewin, 2004). APC-Cdh1 activity
early in G1 verifies that all mitotic proteins are destroyed prior to the initiation of DNA
replication in the next cell cycle (McGowan, 2003; Sullivan and Morgan, 2007).
Once the nuclear envelope is broken-down, checkpoint proteins are recruited to
the outer kinetochore surface of all unattached chromosomes. A single unattached or mis-
attached kinetochore emits an inhibitory diffusible signal to APC/CCdc20, whose nature,
although not yet completely understood, prevents anaphase onset until all chromosomes
are adequately attached and aligned (Bharadwaj and Yu, 2004; Kops et al., 2005; May
and Hardwick, 2006).
- 2 0 -
Checkpoint activated
^ BubR1/Bub3 Bub1/Bub3 ? V CENP-E| t
V ' \ t »
TTK/Mps1 ? MAPK? Plk1 ? Aurora-B ?
Checkpoint silenced
Checkpoint proteins displaced Production of Mad2* ceases (?)
Mad2* —JAPC/CCd Secunn degradation (Lib-dependent)
H (.beparase; » •
Cohesin Cohesin aeavage Anaphase
onset
Cohesin phosphorylation ?
Fig. 1.6: The spindle assembly checkpoint. The kinesin-related protein CENP-E interacts with BubFH and
together they translate structural information about the presence or absence of correctly bound microtubules
at the kinetochore. If there are non-attached or mis-attached chromosomes, these events are thought to
mediate the recruitment of Mad1-Mad2 complexes. Mad2 is then released in a closed conformation (here
represented as Mad2*), strongly inhibiting APC/CCdc20. Securin degradation (normally dependent on
ubiquitination (Ub)) is then prevented, thereby inhibiting the cleavage of cohesin by separase. Sister
chomatids remain bound until proper kinetochore-microtubule interactions are established. Proteins that have
been also found at the kinetochores and whose participation in checkpoint signalling has been reported are
also represented (MAPK, Plk1, Aurora B) (Nigg, 2001).
APC/C inactivation proceeds whenever there are non-attached or mis-attached
chromatids, through Cdc20-mediated recruitment of Mad2, Bub3 and BubR1 (Bub1-
related protein, the vertebrate homolog for yeast Mad3). Together, these proteins
constitute, in near equal stoichiometry (Hoyt, 2001), a quaternary assembly, the mitotic
checkpoint complex (MCC), which acts to sequester and inhibit APC/C regulatory protein
Cdc20 (Bharadwaj and Yu, 2004; Schmidt and Medema, 2006; Logarinho and Bousbaa,
2008). Currently, it is accepted that kinetochores may act as a catalytic platform for MCC
assembly or, alternatively, that they activate checkpoint proteins to form MCC at the
cytosol and to sensitize APC/C for inhibition (Logarinho et al., 2008; Musacchio and
Salmon, 2008).
Kinetochore-associated CENP-E binds to BubP.1, activating its kinase activity,
which is, in turn, necessary for the recruitment of a stable Mad1-Mad2 heterodimer (Kops
et al., 2005; Logarinho et al., 2008). Indeed, fully inactivation of APC/C is reached by the
synergistic interaction between Mad2 and BubR1, which show a lower inhibitory potential
when acting independently (May and Hardwick, 2006). Together with other essential
-21 -
checkpoint proteins, the Mad1-Mad2 heterodimer activates Mad2 which, in conjunction
with BubR1 and Bub3, strongly combine with Cdc20 and prevent it from activating the
APC/C. MCC-promoted sequestration of Cdc20 compromises APC ubiquitin ligase
activity, which is essential for anaphase onset, as explained before (fig. 1.6) (Kops et al.,
2005; Schmidt and Medema, 2006; Logarinho and Bousbaa, 2008). Thus, in normal
situations, only when all the conditions are gathered does the cell progress into anaphase;
otherwise, it remains blocked in the metaphase-to-anaphase transition.
Although controversial, the activation of the mitotic checkpoint response is more
likely to be the result of a combination of both lack of microtubule attachment and of the
consequent lack of physical tension between sister kinetochores (Bharadwaj and Yu,
2004; Pinsky and Biggins, 2005; May and Hardwick, 2006). When all chromosomes
undergo bipolar attachments to spindle microtubules, tension between sister kinetochores
promotes checkpoint silencing and anaphase onset (Zhou et al., 2002; Schmidt and
Medema, 2006). Anaphase inhibitory complexes must then be disassembled and mitotic
checkpoint proteins must be withdrawn from the kinetochores. Dynein and dynactin play a
crucial role in checkpoint protein transport from the kinetochores to the spindle poles
(Howell et al., 2001; Bharadwaj and Yu, 2004). Mad2 stops being recruited to the
kinetochores and the MCC is no longer produced; Cdc20 dissociates from Bub1 and
Mad2 and binds to the APC, rendering it active (Sudakin et al., 2001; Musacchio and
Salmon, 2007). Securin is then degraded, which releases separase for cohesin cleavage.
Cells finally enter anaphase, which is also promoted by cyclin B destruction (Peters, 2002;
Kops etal., 2005; Logarinho and Bousbaa, 2008).
When cells are not capable of satisfying SAC requisites after a long mitotic delay,
they may have different fates: some of them die during mitosis, others exit mitosis but die
via apoptosis in G1, and others exit mitosis but are tetraploid and reproductively dead
(Niikura etal., 2007).
1.4 Checkpoint defects, aneuploidy and cancer
It has been over than 100 years since Theodor Boveri first hypothesized a causal
connection between genome mis-segregation, the disadvantageous effects that it has on
cells and their progeny, and the development of tumour cells (Kops et a/., 2005; Weaver
and Cleveland, 2005). Indeed, genetic instability lies beneath some pathological
disorders, being correlated with premature ageing, inherited diseases and predisposition
to cancer in humans (Aguilera and Gómez-González, 2008).
- 2 2 -
Aneuploidy has generally a dramatic impact over cells, in which, under normal
conditions, the checkpoint response is chronically activated. Cells enter interphase without
cytokinesis, which is normally followed by apoptosis-mediated cell death (Kops et ai,
2005; Schmidt and Medema, 2006). However, tumour cells constitute an exception to this
observation - most solid tumour cells exhibit chromosome rearrangements and are
aneuploid. Different cancer cell lines present 'chromosomal instability' (CIN), which means
that they frequently gain or lose whole chromosomes along with their divisions (Bharadwaj
and Yu, 2004; Kops etal., 2005; Aguilera and Gómez-González, 2008).
Some tumours are stably aneuploid, which implies that, at some point in their
evolution, their cells suffered from a chromosomal redistribution that has somehow
conferred them a proliferative advantage. For instance, in a considerable number of
colorectal cancer cell lines, CIN was detected (Grabsch et al., 2003; Grabsch et ai, 2004;
Abai er ai, 2007) and chromosome content was shown to undergo constant
rearrangements. The exact mechanism that lies behind these chromosomal imbalances is
not precisely established, but one of the most accepted theories postulates that their
origin resides in the combination of defects in apoptosis-related machinery and the
occurrence of failures in the processes that monitor chromosome segregation during
mitosis, one of them being the spindle assembly checkpoint (Kops er ai, 2005; Schmidt
and Medema, 2006).
Furthermore, it is now known that all human aneuploidies (cells that display an
abnormal chromosome content and number, i.e., a number other than 46) that occur
during development end in embryonic lethality or, in the scarce cases in which they do
not, result in serious birth defects. A complete inactivation of certain checkpoint genes in
higher organisms (e.g., Mad2 and Bub1) has the same effect, suggesting that spindle
checkpoint may be crucial for early development of multicellular organisms or that these
genes may have non-checkpoint functions yet to be identified (Bharadwaj and Yu, 2004).
Aneuploidy may result from different phenomena. Aberrant mitotic processes, in
addition to deficiencies in centrosome duplication, maturation and segregation, explain
episodes of divisions occurring in the presence of multipolar mitotic spindles, which, in
turn, generate aneuploid daughter cells. Chromosome cohesion defects may equally
justify aneuploidy appearance in some tumours. A third possibility refers to inappropriate
attachment of chromosomes to spindle microtubules; for instance, merotelic attachments,
in which the same kinetochore is simultaneously connected with microtubules from both
poles, are a common cause of aneuploidy. In colorectal tumours, 90% of which exhibit
CIN, mutations in APC seem to be the main responsible for aneuploidy (Tighe et ai, 2001 ;
Kops er ai, 2005). On one hand, their proliferation is enhanced due to the inability to
degrade (3-catenin; on the other hand, because kinetochore-microtubule interactions are
- 2 3 -
destabilized, chromosomes mis-segregate during anaphase. Lastly, molecular defects
that weaken the mitotic checkpoint response induce aneuploidy by propitiating premature
chromosome segregation (Kops era/., 2005).
Mutations in the genes encoding for checkpoint proteins themselves have also
been identified in some human cancers, although they are rare. For instance, in some
colorectal cancer cell lines, cells presented mutations in one of the Bub1 alleles, and
some other mutations have been thereafter detected in several studies using CIN cancer
cell lines. It is believed that mutations affecting nonidentified spindle checkpoint genes
may also account for other tumours where CIN has been reported (Nigg, 2001 ; Bharadwaj
and Yu, 2004; Grabsch, 2004).
Nevertheless, aneuploid tumour cell lines do not necessarily present mutations in
the known spindle checkpoint genes. In many of them, the observed checkpoint defects
are linked to altered expression levels of spindle checkpoint proteins. Even relatively
minor alterations of this nature can catalyse tumorigenesis, as demonstrated by Michel et
al. (2001) in mice with reduced Mad2 levels that showed a high incidence of lung tumours
(Bharadwaj and Yu, 2004). Moreover, Mad2 promoter silencing has been implicated in
human hepatocellular carcinogenesis; also, reduced Mad2 expression levels, as well as
Bub1 overexpression, have been positively correlated with a deregulated SAC function
and CIN in a breast cancer cell line (Yuan er a/., 2006). Overexpression of Bub1, Bub3 or
BubR1 is frequent among human gastric cancers, in which Mps1 upregulation has also
been documented (Rimkus etal., 2006; Schmidt and Medema, 2006).
The association of the CIN phenotype with mutations in spindle checkpoint genes,
decreased levels of spindle checkpoint proteins and loss or diminishment of the spindle
checkpoint response strongly suggest a close relation between altered activity of the
mitotic checkpoint and tumorigenesis (Niikura era/., 2007).
Since even tumour cells tolerate CIN only to a limited degree (as pronounced
deviations from the normal karyotype might seriously compromise cell viability), it is more
likely that partial disruption of the spindle checkpoint is more commonly observed in these
cells than complete inactivation of the checkpoint response (Bharadwaj and Yu, 2004).
Anyway, it is certain that cell viability-compatible karyotypic abnormalities correlate with
poor prognosis in most cases (Nigg, 2001).
- 2 4 -
1.5 Medulloblastoma and Malignant Peripheral Nerve
Sheath Tumour
Brain tumours are the most common solid tumours during infancy and the second
main form of cancer after leukaemias, constituting also the major cause of oncologic
mortality in paediatrics. Additionally, combined effects of the pathology and of its
treatment represent main morbidity causes (Yazigi-Rivard et ai, 2008).
Medulloblastoma is a highly malignant cerebellar tumour, the most prevalent
malignant brain tumour in childhood. It originates in the cerebellum or posterior fossa,
possibly from cerebellar granule cell precursors (CGNPs) (Uziel era/., 2006; Yazigi-Rivard
et ai, 2008), and is included in the family of cranial primitive neuroectodermal tumours
(PNET). Tumours arise with a similar prevalence in the cerebellar vermis (mainly in
children) and in the cerebellar hemispheres (in older patients), and often invade the fourth
ventricle, with occasional brainstem involvement (Collins, 2004; Yazigi-Rivard et a/.,
2008). Evidence suggests that medulloblastomas may result from CGNPs that
accumulate oncogenic mutations and are not eliminated by cell cycle control mechanisms
that, in turn, depend primarily upon the retinoblastoma protein (Rb) and p53, thereby
failing to correctly exit the cell cycle and differentiating. This way, medulloblastoma cell
precursors are provided (Uziel era/., 2006).
It occurs mostly in children aged 5 to 10 years and accounts for 20-30% of all
paediatric brain tumours (Wechsler-Reya, 2003; von Bueren er ai, 2007; Yazigi-Rivard et
ai, 2008), representing 14.5% of newly diagnosed cases (Gurney et ai, 2000). In adults,
the disease is rather rare, corresponding to less than 2% of all central nervous system
malignancies (Wechsler-Reya, 2003; von Bueren et ai, 2007; Yazigi-Rivard et ai, 2008).
Medulloblastoma has a higher incidence in males (62%) than in females (38%).
Furthermore, it is more common in younger children than in older children; 40% of all
cases are diagnosed before the age of 5, 31% are between the ages of 5, 18.3% between
the ages of 10 and 14, and 12.7% between the ages of 15 and 19 (von Bueren et ai,
2007).
Several lines of evidence suggest that defects in the Sonic hedgehog-Patched
signalling pathway, present in -30% of cases, are implied in the pathogenesis of
medulloblastoma (Wechsler-Reya, 2003; Ferretti et ai, 2005; Uziel et ai, 2006; Yazigi-
Rivard et ai, 2008). Sonic hedgehog (Shh), a molecule secreted by the cerebellar
Purkinje neurons, functions as a master regulator of embryonic development and of
proliferation in the developing cerebellum. Patched (Ptch), in turn, is a transmembrane
protein that acts as both a Shh receptor and an antagonist of its signalling. In effect,
-25-
mutations in the human Ptch gene have been identified in patients with Gorlin's syndrome,
a disease with increased incidence of medulloblastoma. Many sporadic
medulloblastomas, especially those of the desmoplastic subtype, have also been
associated with mutations in Ptch and other components of the Shh-Ptch pathway
(Wechsler-Reya, 2003; Collins, 2004; Pogoriler, 2006).
Mutated elements of the Wnt signalling pathway have been related to
medulloblastoma development as well (Wechsler-Reya, 2003; Uziel et al., 2006). For
instance, Turcot's syndrome, which derives from mutations in the APC gene, is
characterized by a high incidence of colorectal cancers and brain tumours, namely
medulloblastoma (Wechsler-Reya, 2003; Collins, 2004; Pogoriler, 2006). In fact, APC
mutations were reported in about 4% of sporadic medulloblastomas, of which 8-15% have
been shown to harbour activating mutations in B-catenin and 12% in axin, a negative
regulator of Wnt signalling. Thus, a subset of medulloblastomas may derive from
activation of the Wnt pathway (Wechsler-Reya, 2003). Mutations affecting the NOTCH
signal transduction pathway have also been described in medulloblastomas, though
genetic lesions behind the majority of these tumours still remain unidentified (Uziel et al.,
2006).
Several other studies using medulloblastoma cells have proven their
overexpression of transcription factors N-myc, c-myc (Li et al., 2005; von Bueren et al.,
2007), pax5 and zic and of the receptor tyrosine kinase ErbB2 (Wechsler-Reya, 2003;
Collins, 2004). It remains uncertain whether these genes participate in the development
and/or progression of medulloblastomas or if they simply constitute molecular markers of
the transformed cell type (Wechsler-Reya, 2003). The PDGF receptors and ligands are
also being investigated for their eventual significance in medulloblastoma biology (Collins,
2004).
Cytogenetic techniques have revealed that 30 to 50% of human medulloblastomas
carry a deletion or rearrangement of part of chromosome 17 (Wechsler-Reya, 2003; De
Smaele et al., 2004; Ferretti et a/., 2005). In most cases, isochromosome 17q [i(17q)] is
the observed karyotypic abnormality. Most of the short arm (17p) is lost from two
chromosomes 17 and they fuse head-to-head, producing a chromosome with two
centromeres, little 17p and two long arms (17q) (Wechsler-Reya, 2003; Collins, 2004;
Ferretti et al., 2005). This genetic lesion may account for Shh pathway deregulation in the
cases where no mutations in its components are detected (Ferretti et al., 2005). The fact
that this rearrangement is frequently also identified in leukaemias, lymphomas and
stomach, colon and cervix cancers suggests that at least one potent tumour suppressor is
located on chromosome 17p short arm (Wechsler-Reya, 2003). This putative tumour
suppressor is believed to localise to 17p13.3, a region containing about 20 known genes,
-26-
including those encoding the lissencephaly-associated protein Lis1, the breakpoint cluster
region-related gene ABR, the Max-binding protein Mnt, and the transcription factor
hypermethylated in cancer-1 H i d , although none of them have been conclusively linked
to medulloblastoma etiology (Wechsler-Reya, 2003). In this context, it has been asserted
that chromosome 17p deletion leads to the loss of REN(KCTD11), a novel Hedgehog
antagonist, thus abolishing a checkpoint of Shh-dependent events during cerebellum
development and tumorigenesis (De Smaele et al., 2004; Ferretti et al., 2005). Similarly,
p53 tumour suppressor also maps to 17p13 region; its inactivation cooperates with Shh
pathway for medulloblastoma formation, providing haploinsufficiency conditions that cause
abrogation of direct and indirect checkpoints of Shh signalling in cancer (De Smaele et al.,
2004). 10q loss is one of the many other chromosomal aberrations that have been
correlated with medulloblastoma (Collins, 2004).
Malignant peripheral nerve sheath tumour (MPNST) is a rare variant of soft tissue
sarcoma of ectomesenchymal nature. World Health Organization has devised the term
MPNST, allowing the replacement of previous heterogeneous and frequently confusing
terminology, such as malignant schwannoma, malignant neurilemmoma, and
neurofibrosarcoma, for tumours of neurogenic origin and similar biological behaviour (Kar
et ai, 2006). MPNST diagnosis is often made difficult due to the similarities with other
spindle cell sarcomas. The tumour arises from a major or minor peripheral nerve branch
or sheath of peripheral nerve fibres. It may be of spontaneous origin in adult patients,
even if 5 to 42% of MPNST cases are associated with multiple neurofibromatosis Type-I
(Kobayashi er a/., 2006). Multifocality and development of second primary tumours
presenting the same histology are other clinical features of the disease. Surgery is the
preferred treatment for this aggressive tumour (Kar et al., 2006).
Many structural and numerical chromosomal aberrations have been described in
MPNST (Kobayashi et al., 2006). p53 mutations were proven to contribute to the
development of the disease (Berghmans etal., 2005).
- 2 7 -
1.6 Aim of the study
Medulloblastoma is the most common primary central nervous system tumour that
arises in childhood, comprising 20-30% of all paediatric brain tumours. Like in many other
neoplasms, medulloblastoma cells are genetically unstable, reflecting chromosome gain
or loss when cell division takes place. The molecular basis behind medulloblastoma
aneuploidy is not completely known. Nonetheless, structural and numerical karyotypic
abnormalities, as well as mutations in key components of pathways involved in
development and cell proliferation, have been reported in medulloblastoma cells, which
leads to the suspicion that mitotic checkpoint anomalies may contribute to the neoplasm's
pathogenesis and/or malignancy.
In view of medulloblastoma's high representativeness among paediatric tumours
and of the scarcity of studies concerning mitotic checkpoint activity in cells from this
pathology, the present work aims at evaluating the mitotic checkpoint competence in
medulloblastoma cells and identifying the molecular alterations that lie beneath eventual
defects in its activity.
In effect, the amount of works done in this field is far from being satisfactory and
from clarifying the eventual participation of mitotic checkpoint defects in the development
of medulloblastoma tumour cells. Taking that into consideration, the present study aims at
complementing the currently available knowledge concerning medulloblastoma
pathogenesis, namely by characterizing mitotic checkpoint efficiency. Eventual mitotic
checkpoint abnormalities, as well as the underlying molecular alterations, are investigated,
using the Daoy medulloblastoma-derived cell line.
- 2 8 -
2. Materials and Methods
2.1 Cell lines
Three different human tumour cell lines were used: HeLa - cervical cancer -
(Molecular Medicine Institute, University of Lisbon, Portugal), S462 - Malignant Peripheral
Nerve Sheath Tumour - and Daoy - medulloblastoma - (ICVS, University of Minho,
Braga, Portugal).
Although they derive from neoplastic tissue, the efficacy of the mitotic checkpoint
in HeLa cells is well documented, and they constituted the experimental control in all the
experiments.
2.1.1 Cell culture conditions
HeLa, S462 e Daoy cells were cultured in complete growth medium consisting of
Dulbecco's Modified Eagle Medium (D-MEM) supplemented with 10% (v/v) Foetal Bovine
Serum (FBS), 1% (v/v) GlutaMAX and 1% (v/v) Antibiotic-Antimycotic mixture (10,000
units/ml penicillin G sodium, 10,000 pg/ml streptomycin sulfate and 25 pg/ml amphotericin
B as Fungizone®, in 0.85% saline). Unless otherwise stated, all reagents were purchased
from Invitrogen. Cells were cultured in a monolayer, in tissue culture flasks (Orange
Scientific) and kept in exponential growth in a humidified incubator (Heraeus Hera cell) at
37.0 °C and a 5.0% C02 atmosphere. All cell culture-related proceedings were done in a
Nuaire™ laminar flux biological safety cabinet, biosafety level II, type A/B3.
2.1.2 Cell subculture
Cells were splitted before reaching total confluence, in order to avoid suffering
from space and nutrient depletion. Growth medium was aspirated and the cells were
washed with 2 ml of sterile Phosphate-Buffered Saline (PBS, pH 7.2: 137 mM NaCI; 2.7
mM KCI; 6.4 mM K2HP04; 1 mM NaHP04) and incubated with 1 ml of 0.25% Trypsin
EDTA 1x. After a 2 to 3 minute-incubation at 37.0 °C, cell detachment was confirmed by
microscopic observation. 1 volume of complete growth medium was added to stop the
reaction and a 30 pi aliquot of cell suspension was collected. 30 pi of trypan blue were
then added to this aliquot. After a 5 minute-incubation, an aliquot of 10 pi was transferred
to a Neubauer counting chamber. Cells were then counted using the trypan blue exclusion
- 2 9 -
method and cell density (number of cells per ml) was determined1. Finally, a volume
equivalent to 0.2 x 106 cells or to 0.4 x106 cells (unless otherwise stated) was transferred
to a final volume of 5 or 13 ml of fresh growth medium, respectively.
2.1.3 Cell line thawing
Cell lines were stored in Recovery™ - Cell Culture Freezing Medium, containing
10% (v/v) DMSO, in liquid nitrogen. A cryovial containing cells was removed, rapidly
thawed in a 37 °C water bath, and the cells were transferred to pre-warmed growth
medium. This medium was replaced 6 hours later (or, alternatively, the following day) due
to DMSO toxicity, then periodically during culture.
2.1.4 Cell line freezing
Cells were stored when growing exponentially in order to maintain a stock for
posterior thawing and preventing phenotypic degeneration. Cell suspensions were
obtained as mentioned in section 2.1.2 and centrifuged at 1,000 rpm for 5 minutes; the
supernatant was discarded. Each pellet (1.0 to 2.0 x 106 cells) was then gently
resuspended in 1 ml Recovery™ - Cell Culture Freezing Medium containing 10% (v/v)
DMSO, and transferred to adequate cryovials. Upon remaining 15 minutes at room
temperature, the cryovials were transferred to -80 °C for 24-48 hours and finally stored in
liquid nitrogen.
2.2 Cell synchronization
0.08 x 106 cells from each cell line were seeded onto 35 mm individual Petri
dishes (Orange Scientific) and maintained under the conditions that ensure their optimal
growth. When they became about 70% confluent, thymidine (Sigma-Aldrich) was added to
growth media at a final concentration of 2 mM. After 18 to 19 hours of incubation, it was
verified, by phase contrast microscopy in a Nikon Eclipse TE2000-U microscope, that
cells were successfully synchronized in G1/S transition, due to the absence of mitosis.
Cells were then released by removing the culture media, washing three times with PBS
and incubating with thymidine-free growth medium. 11 to 12 hours later, phase contrast
observations confirmed that cells had achieved mitosis.
1 . . . . . . total cell number ... . . .„4 No. of cells/ml = x dilution factor x 10
number of squares
- 3 0 -
2.3 Mitotic index determination
Cells from each cell line were synchronized as detailed in section 2.2. Nocodazole
(Sigma) was added to a final concentration of 0.5 uM three hours after the release. Mitotic
index (Ml) was determined by cell-rounding under phase contrast microscopy (Nikon
Eclipse TE2000-U microscope), as the percentage of mitotic cells in relation to total cell
number (mitotic and interphase cells). Determinations were done upon 0, 9, 15, 21, 27
and 48 hours after the release. Controls were submitted to the same treatment, but with
no nocodazole addition. Three independent determinations were done for each
experimental situation, and a mean value was then calculated.
2.4 Immunofluorescence
For immunofluorescence, cells were seeded in 35 mm Petri dishes (Orange
Scientific) onto 22-mm poly-L-lysine pre-treated glass coverslips two days prior to the
experiment. They were removed from growth medium and fixed either immediately or after
a 3 hour-incubation with 0.5 pM nocodazole (protocol 2). According to the
proteins/structures to detect by fluorescence microscopy in each experience, two variants
of the immunofluorescence protocol were used. For the vast majority of the purposes,
protocol 1 was used, while protocol 2 was followed when preservation of microtubules
was relevant.
Protocol 1. Cells were fixed in freshly prepared 2% (w/v) paraformaldehyde (PFA,
Sigma) in 1x PBS for 7 minutes at room temperature. They were then washed three times
for 5 minutes with 1x PBS. Upon washing, cells were permeabilized by incubating for 7
minutes with 1x PBS containing 0,2% (v/v) Triton X-100. Following permeabilization, cells
were washed three times for 5 minutes with PBST (1x PBS containing 0.05% (v/v)
Tween-20). In order to prevent non-specific binding, cells fixed onto the glass coverslips
were incubated with blocking buffer (PBST containing 10% (v/v) FBS). They were then
labelled with primary antibodies, diluted in PBST containing 5% (v/v) FBS. Different
primary antibodies were employed: rabbit anti-Bubl (1:3000, Abeam), mouse anti-BubR1
(1:600, Chemicon International), mouse anti-Bubl and sheep anti-BubR1 (1:2000, gifts
from S. Taylor, School of Biological Sciences, University of Manchester), human anti-
CREST (1:1500, courtesy of E. Bronze-da-Rocha, IBMC, Porto) and mouse or rabbit anti-
a-tubulin (1:2000, Sigma and Abeam, respectively). Cells were then washed three times
for 5 minutes with PBST. Secondary antibodies were diluted 1:1500 in PBST containing
5% (v/v) FBS. Alexa Fluor® 488- and 568-conjugated secondary antibodies (Molecular
-31 -
Probes, Invitrogen Detection Technologies) were used in accordance with primary
antibodies' sources. Incubations with blocking buffer, primary and secondary antibodies
proceeded within a humidified chamber, for 1 hour at room temperature. Cells were
washed again with PBST three times for 5 minutes and, finally, they were washed once
with 1x PBS for 5 minutes. Each coverslip was mounted in 8 pi Vectashield containing 0.5
pg/ml DAPI and secured with nail polish.
Protocol 2. Cell fixation was done by incubating with methanol for 2 minutes at -
20 °C. Cells were then carefully washed with Tris-Buffered Saline (TBS: 20 mM Tris HCI,
pH 7.4; 150 mM NaCI). Permeabilization consisted of a 5 minute-incubation with TBS
containing 0.5% (v/v) Triton X-100, at room temperature, followed by three 1.5 minute-
washings with TBS containing 0.05% (v/v) Triton X-100. Blocking and antibody dilutions
were done as in protocol 1, except that TBS was used in place of PBS.
Immunofluorescence images were acquired using a 60x apochromatic objective, on a
Nikon Eclipse TE2000-U microscope equipped with a DXM1200F digital camera and
controlled by Nikon ACT-1 software. Representative focus plans were chosen for image
acquisition.
2.5 Chromosome spread
Chromosome spread experiments were performed after 0, 12 and 24 hours of
exposure to nocodazole. Mitotic cells were shaken off and collected into a Falcon tube
(Orange Scientific). In order to promote hypotonic shock, bidistilled water (pre-warmed in
a 37 °C water bath) was added to each cell suspension (60% of the volume of the cell
suspension). The mixture was gently homogenized and incubated for 5 minutes at 37 °C.
It was then centrifuged at 1,000 rpm for 5 minutes. The supernatant was discarded with
the aid of a pipette and the pellet was resuspended in 2 ml of a 75% (v/v) methanol:25%
(v/v) acetic acid solution. Cell fixation in this mixture proceeded for 2 minutes at room
temperature. The suspension was again centrifuged and fixed according to the
abovementioned. The supernatant was discarded again, leaving just 100 to 200 pi of it for
resuspending the pellet with a micropipette. This final suspension was transferred to a
microscope glass slide, dried overnight and was finally mounted in 8 pi Vectashield
containing 0.5 pg/ml DAPI.
- 3 2 -
2.6 Preparation of protein extracts
Growth media in the culture flasks were recovered, cells were washed with PBS
and their detachment was promoted by incubating with Trypsin EDTA for 3 minutes. The
reaction was stopped by adding one volume of growth medium and the resulting
suspensions were combined with the growth media initially collected. The mixtures were
then centrifuged at 1,000 rpm for 5 minutes and the supernatants were discarded. Each
pellet was resuspended in 160 pi RIPA lysis buffer 1x (Santa Cruz Biotechnology), which
was previously supplemented with PMSF, protease inhibitor cocktail and sodium
orthovanadate solutions (also supplied in the product) according to manufacturer
directions. The solution was collected into an ice-cold Eppendorf microcentrifuge tube. In
order to promote cell lysis, the solution was sheared through a syringe-adapted needle.
The lysate was allowed to sit for 20 minutes on ice and was then centrifuged at 13,000
rpm, for 5 minutes and at 4 °C using a Heraeus Biofuge primo R refrigerated centrifuge.
150 U.I of the supernatant were reserved for Western blotting and 10 pi for protein
quantification. Additional 30 pi of 6x SDS-PAGE sample buffer (0.35 M Tris HCI, pH 6,8;
30% (v/v) glycerol; 10% (w/v) SDS; 9.3% (w/v) DTT; 0.25% (w/v) bromophenol blue) were
added to each 150 pi Western blotting sample portion. Then, the tubes were placed on a
heating block and protein samples were boiled at 100 °C for 3 minutes, and then back
onto ice. All samples were frozen at -80 °C.
2.6.1 Protein quantification Protein quantification assays were performed by Optical Density (OD)
spectrophotometry measurements at 562 nm, using the BCA™ Protein Assay Kit (Pierce
Biotechnology), by mixing 50 parts of Reagent A with 1 part of Reagent B and following
the standard test tube procedure protocol. Protein samples were thawed and diluted ten
fold in distilled water before colour development.
2.7 Western blotting
2.7.1 Protein separation by SDS-PAGE Protein samples were analyzed by Western blot, by first separating protein bands
by SDS-PAGE. The gel was formed with 10% polyacrylamide resolving gel (for all the
proteins except for Bub3, which was resolved through a 15% gel) and 6% polyacrylamide
stacking gel. A Mini-PROTEAN® 3 system (Bio-Rad Laboratories) was used for
- 3 3 -
electrophoresis assays. It was filled with running buffer (25 mM Tris HCI, pH 8.3; 192 mM
glycine; 0.1% (w/v) SDS); then, a volume of sample equivalent to 15 ug of protein (except
for Bub3 blots, in which 20 u.g were used) were loaded into the appropriate wells. The
apparatus was run at constant voltage (200 V/gel), for 60 minutes, until the dye front
reached the bottom of the polyacrylamide gel.
2.7.2 Membrane transfer
At the end of electrophoresis, the gels were equilibrated in transfer buffer (192 mM
glycine; 25 mM Tris base; 20% (v/v) methanol, pH 8.3), as well as blotting papers
(Quickdraw™, Sigma-Aldrich) and nitrocellulose transfer membranes (Hybond™-ECL™,
Amersham Biosciences). Gel transfer was done at 110 mA per gel for 75 minutes, using a
Hoefer SemiPhor Transphor Unit (semi-dry transfer unit) from Amersham Pharmacia
Biotech. Voltage was carefully monitored to ensure it did not surpass 25 volts. The
efficiency of the transfer process was assessed by membrane staining with Ponceau S
(0.5% (w/v) Ponceau S; 5% (w/v) trichloroacetic acid).
2.7.3 Blotting
The transferred membranes were washed in Tris-Buffered Saline and Tween-20
(TBST: 137 mM NaCI; 20 mM Tris base; 0.1% (v/v) Tween-20) to remove any remaining
acrylamide. Non-specific sites were blocked with appropriate blocking buffer (5% (w/v)
non-fat dry milk in TBST). The membranes were then washed with a solution of 1% (w/v)
non-fat dry milk in TBST, for 10 minutes at room temperature. Various primary antibodies
were used: mouse anti-Bubl and sheep anti-BubR1 (1:2000, courtesy of S. Taylor,
School of Biological Sciences, University of Manchester), mouse anti-Bub3 (1:1000, BD
Transduction Laboratories) and mouse or rabbit anti-a-tubulin (1:2000, Sigma or Abeam,
respectively). a-Tubulin blots were used as loading controls. Incubations with blocking
buffer, primary and secondary antibodies proceeded for 1 hour each, at room temperature
and with constant agitation. Primary and secondary antibodies were diluted in a solution of
1% (w/v) non-fat dry milk in TBST.
2.7.4 Developing Westerns
After incubation with primary antibody solution, blots were washed in a solution of
1% (w/v) non-fat dry milk in TBST for 10 minutes. They were then incubated with the
appropriate horseradish peroxidase (HRP)-conjugated secondary antibody solutions.
Peroxidase labelled anti-rabbit IgG, anti-mouse IgG (Sigma) and anti-sheep IgG (Abeam)
- 3 4 -
were diluted 1:3000 and employed according to primary antibodies' sources. Membranes
were then washed twice with TBST for 10 minutes and, finally, once with TBS.
For development, the Enhanced Chemiluminescence (ECL) method was
performed; equal parts of Western Lighting Chemiluminescent Reagent Plus oxidizing
solution and enhanced luminol reagent (ECL solutions A and B, respectively) were mixed
after the third wash. Each blot was then immersed in the resulting ECL solution (0.198
mM coumaric acid; 1.25 mM luminol; 0.009% (v/v) H202; 100 mM Tris HCI, pH 8.5) for 3
minutes before being sealed between plastic for development.
Blots were placed in a photo cartridge (Hypercassette™, Amersham Biosciences)
and exposed to Kodak BioMax Light Film (Sigma-Aldrich) in the darkroom, for varying
amounts of time, before development. Autoradiographic films were automatically
developed.
2.7.5 Densitometry For the quantitative analysis of protein expression, films obtained through the ECL
detection system were scanned in a Gel Doc™ XR densitometer and band intensities
were quantified using The Discovery Series™ Quantity One® 1-D analysis software,
version 4.6.1, both from Bio-Rad Laboratories, Inc.
2.8 Real-Time PCR (two-step procedure)
2.8.1 RNA isolation and analysis All RNA-related manipulation precautions were taken during its isolation. Water
and all necessary glassware and autoclavable items were DEPC-treated, while work
surfaces and pipettes were periodically wiped with a 0.5 M NaOH solution and
nonautoclavable plasticware was occasionally rinsed with a 0.1 M NaOH, 1 mM EDTA
solution.
0.8 x 106 cells from each cell line were seeded onto 10 cm2 Petri dish plates
(Nunclon™ A Surface, Nunc) and maintained under optimal growth conditions. When they
reached about 70% confluence, total RNA extracts from HeLa, S462 and Daoy cell lines
were prepared employing PureZOL™ RNA Isolation Reagent (Bio-Rad Laboratories) and
following supplier's instructions concerning cells growing in a monolayer. Growth medium
was aspirated and 1 ml PureZOL™ RNA Isolation Reagent was added to each dish plate.
Up and down pipettings were done in order to promote cell lysis. A 5 minute-incubation at
- 3 5 -
room temperature allowed the complete dissociation of nucleoprotein complexes. Lysates
were collected and 200 (il chloroform were added to each of them. These mixtures were
vigorously shaken for 15 seconds and were then incubated for 5 minutes at room
temperature, with periodical mixings. They were centrifuged at 12,000 g for 15 minutes
and at 4 °C. Upon centrifugation, the (upper) aqueous phases were collected into new
RNase-free tubes, to which 500 pi isopropyl alcohol were added. After thoroughly mixing
and a 5 minute-incubation at room temperature, the mixtures were again centrifuged at
12,000 gr for 10 minutes and at 4 °C. Supematants were carefully discarded and the
resulting pellets were washed with 1 ml 75% (v/v) ethanol, vortexed and finally centrifuged
at 7,500 g for 5 minutes and at 4 °C. Supematants were discarded and pellets were
allowed to air-dry for about 5 minutes, upon which they were resuspended in 50 pi of
RNase-free water, divided into 10 pi aliquots and stored at -80 °C. In quantitative Real-
Time PCR experiments, a total RNA sample from a normal astrocyte cell line, kindly
provided by Rui Reis (ICVS, University of Minho, Portugal), was also included. RNA
quantification was performed by OD measurements at 260 nm. RNase-free deionised
water was used as blank, as well as solvent for sample dilutions. 1 OD unit (260 nm) was
considered to be equivalent to 50 pg/ml RNA. RNA quality was evaluated by
electrophoretic analysis of sample volumes equivalent to 1 pg RNA in 0.8% (w/v) RNase-
free agarose gels, as well as by the OD26onm/ OD28onm ratio.
2.8.2 Quantitative Real-Time Polymerase Chain Reaction
• Primer design
Forward and reverse primers were designed using Beacon Designer 7.2 (Bio-Rad
Laboratories) and PerlPrimer version 1.1.14 (http://perlprimer.sourceforge.net) softwares
(table 2.1) and were obtained from STABvida (Oeiras, Portugal). For subsequent
applications, they were reconstituted in water to a final concentration of 10 pM.
-36-
Gene Primer
Pairing Position
in the cDNA
Length of the
Amplified Sequence
(bp)
Primer Oligonucleotide Sequence
GAPDH Forward 49
101 5'-ACA GTC AGC CGC ATC TTC-3'
GAPDH Reverse 149
101 5'-GCC CAA TAC GAC CAA ATC C-3'
Cdc20 Forward 421
151 5'-CCA GAG GGT TAT CAG AAC AG-3'
Cdc20 Reverse 571
151 5'-CCA CAA GGT TCA GGT AAT AGT-3'
Bub1 Forward 463
171 5'-CAA GCT AGA ACC TCA GAA CC-3'
Bub1 Reverse 633
171 5'-CAC TCT TCG TTC CAT ATT TGA C-3'
BubR1 Forward 512
82 5'-CAG ATG CGA TAT TTC AGG AAG GG-3'
BubR1 Reverse 593
82 5'-GCT TGG AAT TGT CGG TGC TG-3'
Bub3 Forward 475
117 5'-GTG TTG GTG TGG GAC TTA CG-3'
Bub3 Reverse 591
117 5'-GCT TAA TAC ATA ACC CTG CTT G-3'
Mad2 Forward 205
117 5'-GTG GAA CAA CTG AAA GAT TGG T-3'
Mad2 Reverse 321
117 5'-GTC ACA CTC AAT ATC AAA CTG C-3'
Chfr Forward 1671
105 5'-AGACATCCTGAAGAATTACCT-3'
Chfr Reverse 1775
105
5'-TAATCAGACAGCAGAAACACTC-3'
Table 2.1: Sequences of forward and reverse primers used in Real-Time PCR amplification of GAPDH,
Cdc20, Bub1, BubR1, Bub3, Mad2 and Chfr genes using cDNA samples from HeLa, astrocyte, S462 and
Daoy cell lines.
Prior to their use in Real-Time PCR experiments, a Reverse Transcriptase PCR
with isolated HeLa RNA samples was done to confirm absence of genomic DNA
amplification for each primer pair. Then, another PCR was run in order to ascertain primer
specificity and annealing temperatures. Superscript™ III One-Step RT-PCR with
Platinum® Taq DNA Polymerase kit (Invitrogen) was used and reaction mixtures contained
2 pi total HeLa RNA, 25 pi 2x Reaction Mix, 2 pi forward primer, 2 pi reverse primer, 2 pi
Superscript™ RT/Platinum® Taq mix and 17 pi autoclaved distilled water. The
amplification program, run in a MJ Mini™ Thermal Cycler (Bio-Rad Laboratories),
comprised the following steps: 50.0 °C for 20 minutes; 94.0 °C for 1.5 minutes; 94.0 °C for
- 3 7 -
15 seconds; 60.0 °C for 30 seconds; 72.0 °C for 1 minute. Amplification products were
finally run in a 1% (w/v) agarose gel.
• cDNA synthesis
cDNA synthesis was achieved through a reverse transcriptase-mediated reaction,
using the iScript™ cDNA Synthesis Kit (Bio-Rad). 1 pg total RNA from each cell line was
used as template. Each reaction was composed of 4 pi 5x iScript Reaction Mix, 1 pi
iScript Reverse Transcriptase, a volume of RNA template equivalent to 1 pg RNA and
nuclease-free water to a final volume of 20 pi. The program consisted of the steps that
follow: 25.0 °C for 5 minutes; 42.0 °C for 30 minutes; 85.0 °C for 5 minutes. It was run in a
MJ Mini™ Thermal Cycler (Bio-Rad Laboratories). Resulting cDNA samples were stored
at -20 °C.
• Real- Time PCR assays
Each cDNA sample was serially diluted 1:1, 1:10, 1:100 and 1:1000 in RNase-free
water for standard curve plotting. For each Real-Time PCR run, a master mix was
prepared using iQ™ SYBR® Green Supermix kit (Bio-Rad), following its recommendations
and containing 100 nM of each primer, 1x iQ SYBR Green Supermix, and 1 pg cDNA in a
final volume of 25 pi. RNA template controls, as well as non-template controls (in which
sterile deionised water was used in place of template cDNA) were used to confirm
absence of genomic DNA and cDNA contamination, respectively. Experiments were done
in triplicate for each data point. The thermal cycling conditions comprised an initial
denaturation step at 95.0 °C for 3 minutes, 35 cycles at 94.0 °C for 20 seconds, 60.0 °C
for 30 seconds and 72.0 °C for 30 seconds. Finally, the melt curve comprised
temperatures from 65.0 to 95.0 °C, with increments of 0.5 °C, for 5 seconds. The program
was run in a C1000™ Thermal Cycler, interfaced to a CFX96™ Real-Time PCR Detection
System, both from Bio-Rad Laboratories. In the end of each program, data were acquired
and analysed using CFX Manager™ Software, version 1.0 (Bio-Rad Laboratories).
Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) housekeeping gene amplification
was used as a control for normalization of gene expression levels. ACT (ACT = Gene CT -
GAPDH CT) was determined. Differences in expression levels between HeLa, astrocyte
and Daoy and S462 cell lines were calculated using A(ACT) [A(ACT) = HeLa, S462 or Daoy
ACT - astrocyte ACT]. Melting curve analysis was done in order to verify primer specificity.
Real-Time PCR products were finally run in 1% (w/v) agarose gels in order to verify band
intensities and the existence of single amplification products.
- 38 -
2.9 Statistical analysis
Data were statistically analysed by means of the t-Student's test. p-values<0.05 were considered statistically significant at the 90% confidence level, and p-values<0.01 at the 99% confidence level.
- 3 9 -
3. Results and Discussion
The spindle assembly checkpoint or mitotic checkpoint is a surveillance
mechanism that acts at the metaphase-to-anaphase transition by monitoring kinetochore-
to-microtubule attachments. It delays anaphase onset if kinetochore-to-microtubule
attachments are absent or incorrect, thus ensuring appropriate chromosome segregation
and the fidelity of genetic transmission (Rieder et al., 1994; May and Hardwick, 2006;
Logarinho and Bousbaa, 2008).
Medulloblastoma is a highly frequent central nervous system neoplasm in
childhood, representing -20% of all paediatric brain tumours and affecting mostly children
under 5 years old (Wechsler-Reya, 2003; von Bueren et al., 2007; Yazigi-Rivard er ai,
2008). Medulloblastoma cells present characteristic structural and numerical chromosome
aberrations, as well as mutations in important elements of the Shh-Ptch and Wnt
pathways, implied in development and cell proliferation (Wechsler-Reya, 2003; Ferretti et
al., 2005; Uziel et al., 2006; Yazigi-Rivard et ai, 2008). Chromosomal instability, as a
result of abnormal chromosome segregation, may derive from a defective mitotic
checkpoint activity, which is investigated in this work.
The present study aims at analysing: (i) the competence of the mitotic checkpoint,
and (ii) the molecular alterations underlying the eventually defective checkpoint in the
medulloblastoma cell line Daoy.
3.1 The mitotic checkpoint is defective in Daoy cell lines
In order to evaluate Daoy cells' ability to arrest in mitosis when exposed to mitotic
stress-inducing conditions, and thus the efficacy of their mitotic checkpoint, cells were
synchronized, released and then exposed to nocodazole, a microtubule-disrupting drug.
Cell phenotypes were analysed and the mitotic index (the percentage of mitotic cells
relative to the total number of viable cells in culture) was determined after different periods
of incubation with nocodazole. In the same context, chromosome spread assays were
performed so as to ascertain whether sister chromatids in Daoy cells remain bound when
exposed to nocodazole, as expected in case of having a functional and efficient mitotic
checkpoint.
- 4 0 -
In a first approach, the cells' response to the incubation with microtubule poisons
was evaluated. In order to make certain that nocodazole could be used as a cell cycle
disruptor in subsequent experiments, synchronized Daoy, S462 and HeLa cells were
incubated for three hours with 0.5 pM nocodazole. Nocodazole is an anti-mitotic agent
that disrupts microtubules by binding to (3-tubulin and preventing formation of one of the
two interchain disulfide linkages, thus inhibiting microtubule dynamics, disrupting the
mitotic spindle function, and fragmenting the Golgi complex. Because it interferes with
microtubule polymerisation, it impedes mitotic spindle assembly and, therefore, there is no
attachment between kinetochores and microtubules. This lack of attachment induces
chronic mitotic checkpoint activation and the cell cycle is arrested at G2/M phase.
Prolonged contact with nocodazole induces apoptosis in several normal and tumour cell
lines. Nevertheless, some cells may metabolically reject the drug, an hypothesis that must
then be tested.
Upon exposure to nocodazole, cells were immunostained for microtubule structure
visualization. For that, they were stained with an anti-a-tubulin antibody, for the
assessment of the effect of nocodazole on microtubules (fig. 3.1).
-41 -
HeLa S462
Tubulin DAPI + CREST + Tubulin Tubulin DAPI +CREST +
Tubulin
(A co
JZ a.
c
DC h-O
ΠI -O
P^&fl
B H B E3 W^
Daoy
cr CD i -0) CO o
£ a i _
a> +* c
o O 2 +
DC 1-
(A U Cfl O +-
S O O z +
Tubulin Tubulin
H ■ B H D H
Fig. 3.1: Effects of nocodazole exposure in HeLa, S462 and Daoy interphasic and mitotic cells. In the controls
(CTR), no nocodazole was added. Interphasic and mitotic cells were observed under fluorescence
microscopy. Independently of the cell cycle phase, microtubules appear correctly assembled in the absence of
the drug, while in its presence they are depolymerised. DNA is shown in blue, a-tubulin in green and CREST
in red. Scale bars, 5 urn
- 4 2 -
As shown in figure 3.1, in the absence of nocodazole (CTR), interphase cells
show a well-defined microtubule net and these compose a functional, bipolar spindle in
mitotic cells, as expected. However, microtubules become depolymerised in the three cell
lines after nocodazole treatment. These observations indicate that cells metabolise the
drug and that any change in their behaviour upon exposure to it is not due to any kind of
nocodazole rejection mechanism. As a result of nocodazole action, mitotic spindle
assembly is prevented; chromosome capture and congression (and, consequently,
alignment at the metaphase plate) are thus inhibited. Mitotic progression is compromised
and cell division can no longer take place.
3.1.1 Daoy cell line exhibits a low mitotic index upon nocodazole exposure
Because microtubule depolymerisation compromises spindle assembly and
generates unattached kinetochores that are sensed by the mitotic checkpoint, and once it
was shown to occur in Daoy cells when these are exposed to nocodazole, incubation with
the drug was expected to induce a mitotic arrest and, consequently, an increase in mitotic
index. Cells were therefore incubated with nocodazole for distinct periods of time. In each
of them, mitotic index was determined in three independent experiments. The same
procedure was done in the control cultures, but with no nocodazole addition. The mean
values are depicted in fig. 3.2.
Ml upon nocodazole exposure
100 -
80 - - ♦ - H e L a C T R
? 60 ^ 50 I 40
30
/ " - * —*—HeLa + Noc ? 60 ^ 50 I 40
30
/I - * -S462CTR ? 60 ^ 50 I 40
30
/1 , , - * - S 4 6 2 + Noe
? 60 ^ 50 I 40
30 // ^ ^ ■ « v ^ V —•—Daoy CTR
20 10
~"~4gJÍSL — r^^^ —•— Daoy+Noc 20 10
—•— Daoy+Noc 20 10
—•— Daoy+Noc
0 9 15 21 27 48
Time after the release (h)
Fig. 3.2: Mitotic index (Ml) from HeLa, S462 and Daoy cell cultures treated with nocodazole (+ Noc). In the
controls (CTR), no nocodazole was added. Data are mean ± S.D. (n = 3 independent experiments). * p<0.05;
**p<0.01.
- 4 3 -
When compared with control HeLa cells, Daoy control cells show slightly higher
(but comparable) mitotic index values (-7% versus -5%), except for the immediate hours
after the release, in which there is a greater percentage of HeLa mitotic cells (-20%
versus -7%). S462 cells, in turn, are those that exhibit the highest mitotic index values in
the absence of nocodazole (-10%), also achieving a peak value in the first 9 hours of the
release (-17%). The differences in mitotic index values observed between the three
control situations may be justified by their distinct proveniences (HeLa, cervical cancer;
Daoy, medulloblastoma; S462, malignant peripheral nerve sheath tumour).
Concerning the situations where nocodazole was added, it can be stated that
HeLa and S462 cells accumulate in mitosis, as expected considering nocodazole action,
and eventually die as a consequence of a prolonged arrest in prometaphase. In effect, a
pronounced increase in the number of mitotic cells is seen early in response to
nocodazole addition, with mitotic index values rising in the first 6 hours of incubation and
peaking about 15 hours after the release (that is to say, 12 hours upon nocodazole
addition), both presenting 80% mitotic cells. Mitotic index values close to this were again
obtained upon 19 and 24 hours of exposure but were no longer determined from this
moment on because, at about this time, these cells begin to die. The results are in
agreement with the known efficient mitotic checkpoint in HeLa cells and show that S462
cells also have an efficient mitotic checkpoint.
On the contrary, although Daoy cells do accumulate in mitosis - as it can be
deduced from the Ml peak value reached at about 15 hours after the release (36.1%, a
statistically significant different value from the control culture) - , this increase in their
mitotic index is much more subtle than in HeLa and S462 cases (which achieve -85% and
70% mitotic index values, respectively). Then, instead of stabilising its mitotic values and
eventually dying, the number of Daoy mitotic cells actually decreases because they
escape mitosis - as confirmed by the presence of many normal looking interphase cells
(fig. 3.3) - , even after prolonged incubation with the drug. In fact, after 45 hours of
exposure, cells at different cell cycle phases are observed (data not shown). We observed
few Daoy dead cells in the presence of nocodazole, which is consistent with the presence
of a defective mitotic checkpoint in this cell line.
- 4 4 -
Fig. 3.3: Nocodazole effects along with time in synchronized HeLa (A), S462 (B) and Daoy (C) cell cultures.
Phase contrast images (40x magnification) show cell cultures upon a 12 and a 24-hour exposure period to
nocodazole (Noc), as well as the corresponding controls (CTR). Incubation with nocodazole results in
significant differences in the number of mitotic and interphasic cells between the three cell lines. Upon 24
hours of incubation, while HeLa and S462 cultures are arrested in mitosis and start to evidence signals of cell
death, Daoy cultures present colonies of interphasic cells, with a low number of mitotic cells.
3.1.2 Chromosome spread assays
In order to verify to which extent sister chromatids remained held together during
the prolonged mitotic arrest, chromosome spread assays were performed (Maraldi et al.,
1999). As shown in fig. 3.4A and as expected, HeLa and S462 sister chromatids remain
held together during nocodazole-induced mitotic arrest. Nonetheless, Daoy cells show a
high number of mitosis-arrested cells with sister chromatid separation (SCS).
- 4 5 -
CD
I
(O
C/)
> o CO
o
Fig. 3.4A: Chromosome spread results in HeLa, S462 and Daoy cell lines upon an 8 hour-exposure to 0.5 uM
nocodazole. In the presence of the drug, sister chromatids remain held together at the centromere in HeLa
and S462 mitosis-arrested cells. On the contrary, there is a significant number of mitotic cells showing
separated sister chromatids (white arrows) in the Daoy cell line. A magnification of a representative
chromosome is provided for each cell line. Scale bars, 5 urn.
- 4 6 -
o o w
SCS upon nocodazole exposure
■ HeLaCTR
■ HeLa + Noc
"S462CTR
■ S462 + NOC
wDaoyCTR
■ Daoy + Noc
Fig. 3.4B: Percentage of sister chromatid separation (% SCS) in HeLa, S462 and Daoy cell cultures upon a 0,
12- and 24-hour period of incubation with 0.5 uM nocodazole (+ Noc). Data are mean ± S.D. (n = 3
independent experiments). * p<0.05; ** p<0.01.
Fig. 3.4B depicts that, in HeLa and in S462 cells, a very low number of mitotic
cells presented sister chromatid separation (SCS) in the absence of nocodazole (1.9% in
HeLa and 2.7% in S462 cells), which also did not increase significantly in the presence of
the drug upon 12 hours of exposure (7.3% and 5.8% of SCS in HeLa and in S462 cells,
respectively).
On the contrary, Daoy cells showed a steep increase in the percentage of cells
with SCS during incubation with nocodazole. At Oh, Daoy SCS percentage values were
comparable to those from HeLa and S462 cell lines (2.0%), but the frequency of cells with
SCS rose remarkably in the presence of nocodazole (59.5% and 85.8% of the cells
showed SCS after 12 and 24 hours of incubation with the drug, respectively).
Therefore, in spite of being arrested in mitosis, as judged by phase contrast
microscopy observation, a high percentage of Daoy cells presents sister chromatid
separation. These results are consistent with a defective mitotic checkpoint in Daoy cells.
- 4 7 -
3.2 Mitotic checkpoint proteins exhibit a normal subcellular distribution in the Daoy cell line
Once spindle assembly checkpoint was proven to be defective in Daoy cells, the
study was directed towards the analysis of specific checkpoint proteins, so as to establish
a molecular explanation for the observed defects. In a first approach, in order to
qualitatively verify protein cellular localisation and to monitor it along with mitosis
progression, different mitotic proteins were labelled by immunofluorescence in HeLa,
S462 and Daoy cell cultures. Fig. 3.5 and fig. 3.6, respectively, show Bub1 and BubR1
subcellular localisations in different mitosis stages.
- 4 8 -
A HeLa DAP I Bub1 DAPI + Bub1
V) ro .c Q.
0 E o
m to
JZ Q. « Q>
<D (0 CO
a (0 c <
B Daoy DAPI Bub1 DAPI + Bub1
in ro j = a. o
</> ro Q. ro E o
o en ro a ro c <
Fig. 3.5: Bub1 intracellular distribution patterns in HeLa (A) and Daoy (B) cells throughout mitosis. Bub1 is shown in red. DNA was stained with DAPI (blue). Immunofluorescence images show that Bub1 concentrates at kinetochores during prophase and prometaphase and decreases significantly at metaphase and anaphase. The same spatial and temporal distribution is observed in both cell lines. Scale bars, 5 urn.
- 49 -
A HeLa
DAP I BubR1 DAPI + BubR1
a> in ro si Q. O
CD in re
CD in ro .c a. ro a! 2
o in ro a. ro c <
B Daoy DAPI BubFM DAPI + BubR1
0) en ro .c a o
a> <n ro £ Q. ro a! E o et
(D in ro x: a. ro ai
0) en ro -C a m c <
Fig. 3.6: BubR1 intracellular distribution patterns in HeLa (A) and Daoy (B) cells throughout mitosis. BubR1 is shown in red. DNA was stained with DAPI (blue). Immunofluorescence images show that BubFH concentrates at kinetochores during prophase and prometaphase and decreases significantly at metaphase and anaphase. The same spatial and temporal distribution is observed in both cell lines. Scale bars, 5 urn.
- 5 0 -
For these proteins, the same subcellular and temporal distribution patterns were
observed. In both cell lines, as well as in S462 cells (data not shown), this distribution is in
accordance with the reported for these proteins. In effect, Bub1 and BubFM were shown to
localise to kinetochores, to accumulate during prophase and prometaphase, to have a
diminished signal in metaphase and to be undetectable in anaphase, which coincides with
their predictable temporal behaviour while mitotic proteins (Clarke and Giménez-Abián,
2000).
The fact that no detectable changes were observed between control cell lines and
Daoy cells suggests that eventual alterations in spatial and temporal distribution of mitotic
checkpoint proteins do not lie beneath the evident defects of its activity.
3.3 Daoy cell line shows altered expression of mitotic checkpoint genes
Besides analysing the intracellular and temporal distribution of mitotic checkpoint
proteins, the present work also focused on the study of their expression, at the gene and
protein levels, in order to find a molecular explanation for the observed mitotic checkpoint
defects. To do so, protein expression levels were assessed by Western blotting and the
corresponding gene expression levels were evaluated by quantitative Real-Time PCR.
In Western blotting assays, total protein extracts prepared from HeLa, S462 and
Daoy cell cultures exposed to 0.5 pM nocodazole for 16 hours were used along with the
corresponding controls. Bub1, BubR1, Bub3 and a-tubulin were labelled; a-tubulin was
used as loading control (fig. 3.7).
-51 -
Cell line HeLa S462 Daoy Molecular weight (kDa)
Daoy protein expression levels (%)
Nocodazole - + - + - +
Molecular weight (kDa) Relative to
HeLa Relative to
S462
Bub1 ^MNf J ^ ^ B WÈÊÈ W^ÊÊ - i*** iMMMHl T ^ ^ ^ ^ M ^ ^ ^ H V :nv^y>*TW* ^SS^m!^^
Wf 1W 122 85 65
a-Tubulin 55
BubR1 Blx*^b 120 90 82 BubR1 120 90 82
a-Tubulin — 55
Bub3 . - 37 87 98
a-Tubulin 55
Fig. 3.7: Western blots concerning Bub1, BubR1 and Bub3 protein expression levels in HeLa, S462 and Daoy
cell lines growing in the presence (+) or in the absence (-) of 0.5 uM nocodazole for 16 hours. Total protein
extracts were harvested, separated on 10% (Bub1 and BuR1) or 15% SDS-PAGE (Bub3), transferred to
nitrocellulose membranes and immunoblotted using anti-Bubl, anti-BubR1, anti-Bub3 and anti-a-tubulin
antibodies. a-Tubulin was used as loading control. Bub1, BubR1 and Bub3 protein levels in cell cultures
growing in the absence of nocodazole were determined by densitometry and expressed relative to HeLa and
S462 control cultures.
As far as Bub3 protein expression levels are concerned, there were no significant
changes between the three cell lines, with the intensities of a-tubulin-corresponding bands
dissuading any apparent differences.
In turn, analysis of the Bub1 and BubR1 blots reveals that, in HeLa and S462 cell
lines, the levels of expression of these proteins are higher, either in the presence or in the
absence of nocodazole. Furthermore, it must be emphasized that, in the situations where
the drug was added, Bub1 and BubR1 bands are more intense in HeLa and S462, which
was already expected since these are mitotic proteins, thereby accumulating in conditions
in which a mitotic arrest is induced. Since these cells have a very efficient response to
nocodazole and markedly arrest at mitosis, their amount is consequently higher in HeLa
and S462 cells than in those from the Daoy cell line, which accumulate less in mitosis. In
fact, Daoy lower levels of expression of Bub1 and BubR1 as a consequence of a weaker
mitotic arrest in the presence of the microtubule disruptor corroborate the results obtained
in mitotic index determinations, which have shown that these cells accumulate in mitosis
-52-
to a lesser extent when compared with those from HeLa and S462 lines in the same
conditions.
Differences in band intensities of mitotic checkpoint proteins found through
Western blotting assays led to the hypothesis that their expression levels are altered in
Daoy cells. In order to confirm this suspicion, their corresponding mRNA levels were
assessed through Real-Time PCR. Mad2, Bub1, BubR1, Bub3, Cdc20 and Chfr gene
expression levels were studied in this context.
Gene expression analysis by means of Real-Time PCR made use of appropriate
primer pairs, whose specificity was confirmed in a preliminary PCR assay. This
experiment has proven that each primer pair amplified a single fragment. Total RNA
samples were isolated from HeLa, S462 and Daoy cell lines growing in the absence of
nocodazole; an additional astrocyte total RNA sample, prepared in the same conditions,
was also included in the assays. Astrocyte expression levels were used as controls for the
normalization of gene expression since astrocytes are normal glial cells, derived from the
brain and spinal cord, having thus a closer provenience to Daoy cells than those from
HeLa (which derive from cervical cancer). The use of astrocyte expression levels as
controls eliminates the differences in gene expression arising from distinct cell line origins.
Also, gene expression levels were normalized against those of GAPDH, a housekeeping
gene whose expression did not vary between the cell lines used under our experimental
conditions.
Analysis of Real-Time PCR results reveals statistically significant differences in the
expression levels of all the genes that were tested except for Bub1. In effect, Mad2,
BubR1 and Bub3 genes are underexpressed in Daoy cells, whilst Cdc20 is overexpressed
(fig. 3.8).
In Daoy cells, Bub1 gene was found to be only 5% less expressed than in
astrocytes, a little divergence that is not statistically different and, therefore, not
considered as an underexpression. In fact, all four cell lines showed comparable values of
Bub1 expression levels. However, if compared to those of HeLa, Daoy Bub1 expression
levels are more underexpressed, in accordance to what has been observed in the
analysis of protein expression through Western blotting (fig. 3.7).
- 5 3 -
Relative expression of mitotic checkpoint genes 3,00
u Astrocytes
■ HeLa
■ Daoy
US462
Mad2 Bubl EiibRI Bub3
Target gene
Cdc20
Fig. 3.8: Relative mean expression levels of Mad2, Bub1, BubR1, Bub3 and Cdc20 genes in astrocyte, HeLa,
Daoy and S462 cell lines. Expression levels were normalized against those of the housekeeping gene
GAPDH; the resulting values were normalized against those from astrocytes, which were established as 1.00
(100%). Mad2, BubR1 and Bub3 genes are underexpressed in Daoy cells, while that of Cdc20 is
overexpressed. Data are mean ± S.D. {n - 3 independent experiments). * p<0.05; ** p<0.01.
Mad2 encoding gene, in turn, was found to be pronouncedly underexpressed, with
an 80% reduction in its expression levels relative to astrocytes. Similarly, Daoy cell line
was demonstrated to express only 25% and 23% of BubR1 and Bub3 astrocyte levels
(defined as 100%), respectively. Again, if compared with HeLa expression levels, BubR1
is underexpressed in the Daoy cell line, in compliance with what has been observed when
the corresponding protein levels were studied by means of Western blotting (fig. 3.7). A
relevant finding was that S462 cells have a 46% higher BubFH gene expression level
relative to that of HeLa cells, an overexpression that is also substantiated by the
corresponding Western blotting results (fig. 3.7).
On the other hand, Cdc20 gene was shown to be overexpressed in Daoy cells,
where its expression levels are equivalent to 261% of those of astrocytes. High levels of
Cdc20, the essential component for APC/C activity, contribute to premature metaphase-
to-anaphase transitions, propitiating the occurrence of abnormal chromosome segregation
and leading to aneuploidy (Schmidt and Medema, 2006), which may explain why Daoy
cells prematurely exit from mitosis even in the presence of the mictotubule-depolymerising
drug.
Besides investigating the expression levels of genes encoding for mitotic
checkpoint proteins that act specifically at the metaphase-to-anaphase transition, the
- 5 4 -
present study did also comprise the analysis of Chfr gene expression (fig. 3.9). Although it
was also shown to play a role later in mitosis, having been implied in events such as
spindle formation, chromosome segregation, and even in the mitotic checkpoint, Chfr was
primordially reported to intervene in late G2, delaying prophase in case of mitotic stress
(Scolnick and Halazonetis, 2000; Chaturvedi et ai, 2002; Gong, 2004; Loring étal., 2008;
Privette and Petty, 2008). It therefore constitutes an important control point for the cell
cycle to advance into mitosis. Real-Time PCR assays demonstrated an overexpression of
Chfr gene in Daoy cells, in which it is 9.52 times more expressed than in astrocytes. In the
S462 cell line, this gene is 12 times more expressed than in astrocytes (fig. 3.9).
Relative expression of the Chfr gene
I o
O
14,00
12,00
10,00
8,00
6,00
4,00
2,00
0,00
y Astrocytes
"HeLa BDaoy
"S462
Chfr
Target gene
Fig. 3.9: Relative mean expression levels of the Chfr gene in astrocyte, HeLa, Daoy and S462 cell lines.
Expression levels were normalized against those of the housekeeping gene GAPDH; the resulting values
were normalized against those from astrocytes, which were established as 1.00 (100%). Chfr is
overexpressed in Daoy cells. Data are mean ± S.D. (n-3 independent experiments). ** p<0.01.
Taken together, the results demonstrate that, in spite of being present in these
cells, the spindle assembly checkpoint works defectively. A compensatory molecular
mechanism, through which overexpression of some proteins may compensate for the
underexpression of others, may constitute the reason why it partially maintains its function
without compromising tumour cell viability. For instance, Chfr gene overexpression may
represent a compensatory mechanism responsible for keeping genetically abnormal cells,
which result from failures in the mitotic checkpoint activity, from advancing into another
mitosis.
Loss-of-function mutations of genes of the spindle checkpoint machinery are rare
in solid tumours; other mechanisms such as deregulated expression are speculated to
underlie chromosomal instability (Hernando era/., 2001; Saeki et ai, 2002; Rimkus et ai,
- 5 5 -
2004; Schmidt and Medema, 2006). Given mitotic checkpoint proteins' crucial role for
proper mitotic divisions - and, among these, Bub and Mad family proteins' role in
particular - , any alteration in their expression levels may seriously compromise their
functions and, therefore, an adequate chromosomal segregation (May and Hardwick,
2006; Musacchio and Salmon, 2007).
The alterations that were detected in mitotic gene and protein expression may
explain, at least in part, mitotic checkpoint inefficiency observed in the Daoy cell line. In
addition, they may account for medulloblastoma pathogenesis and contribute to its
proliferation and malignancy.
- 5 6 -
4. Conclusions
Sister chromatid segregation is a complex process whose accuracy, as a
prerequisite for genetic stability, is dependent on the activity of the mitotic checkpoint. In
particular, Bub and Mad family proteins have been shown to be crucial for mitotic
checkpoint functions (May and Hardwick, 2006; Musacchio and Salmon, 2007). Defects in
its mechanism are responsible for unequal chromosome partitioning between daughter
cells and, consequently, for loss or gain of chromosomes during cell divisions. The
resulting genetic imbalance, known as aneuploidy, is a hallmark of cancer cells
(Bharadwaj and Yu, 2004; Kops et al., 2005; Sengupta et al., 2007). Although mutations
in genes encoding for checkpoint proteins are rare, alterations in their expression levels
have been reported in several works and are thought to be the actual cause of karyotypic
abnormalities observed in cancer cells (Hernando et al., 2001; Saeki era/., 2002; Rimkus
etal., 2004; Schmidt and Medema, 2006).
Medulloblastoma, the most frequent brain tumour in childhood, is a highly
malignant neoplasm with a poor prognosis in the vast majority of cases. Current treatment
methods combine surgical resection and chemotherapy, which impair physical and
cognitive development. Besides presenting chromosome structural and numerical
aberrations (Wechsler-Reya, 2003; De Smaele et al., 2004; Ferretti et a/., 2005),
mutations in elements of the Shh-Ptch and Wnt pathways, both involved in cell
development and proliferation, are documented in different works with medulloblastoma
cells (Wechsler-Reya, 2003; Collins, 2004; Pogoriler, 2006).
The present work aimed at studying the competence of the mitotic checkpoint in a
medulloblastoma cell line - Daoy - , as well at characterizing the subjacent molecular
alterations.
Our results show that: (1) Daoy cells do have a mitotic checkpoint, but it works
defectively, as demonstrated by their low mitotic index and their ability to escape mitosis
without apoptosis upon incubation with microtubule-disrupting drugs; (2) The mitotic
checkpoint proteins have a normal temporal and spatial distribution in Daoy cells; (3) The
mitotic checkpoint genes have altered expression both at protein and mRNA levels. More
specifically, BubR1, Bub3 and Mad2 were shown to be underexpressed, while Cdc20 and
Chfr were shown to be overexpressed.
- 57 -
Although the mechanisms underlying improper chromosomal segregation and
aneuploidy are not yet completely established, our results have demonstrated that mitotic
checkpoint competence is compromised in the medulloblastoma cell line under analysis
and that alterations in the expression of mitotic checkpoint genes may lie behind the
observed checkpoint inefficiency and contribute to medulloblastoma pathogenesis and/or
progression. A complete understanding of the mitotic checkpoint mechanism and of the
molecular basis underlying its defects in medulloblastomas may hold the key for a novel
strategy for their therapy.
- 5 8 -
5. Future Prospects
It would be suitable to include, in all experimental studies, a primary, normal cell
line having the same origin as that of medulloblastomas, an internal control that would
allow a more direct comparison of the results and legitimate a more immediate validation
of those obtained with Daoy cells. This strategy has actually been followed in the Real-
Time PCR experiments, where RNA from normal astrocytes has been used.
Nevertheless, because the astrocyte cell line itself was not available in the laboratory, it
was not possible to contemplate it in the phenotypic studies presented for HeLa, S462
and Daoy cell lines. Furthermore, it would be interesting to include astrocyte protein
extracts in the Western blotting assays, in order to ascertain whether protein expression
levels reflect gene expression tendencies observed in Real-Time PCR results.
In addition, since Real-Time PCR and Western blotting assays have consistently
shown that, in Daoy cells, some spindle assembly checkpoint genes are underexpressed,
while others are overexpressed, it would be interesting to interfere with the expression of
these genes so as to study the corresponding effects on cell proliferative activity and
survival. Cell line transfection with expression recombinant vectors would allow the
evaluation of protein overexpression effects on mitotic checkpoint activity in the tumour
cell lines under study and the analysis of the overexpression of some proteins as a means
to compensate for the underexpression of others. Conversely, expression levels of
overexpressed proteins could be lowered by means of RNAi-mediated gene knockdown in
order to understand how important the overexpression of these genes is to tumour cell
proliferation and survival.
Since mitotic checkpoint proteins' functions are dependent on the interactions they
establish with each other and with other proteins, this study should be broadened in order
to include a greater number of mitotic checkpoint-related proteins, aiming at characterizing
their distribution and expression in the cell lines under analysis. This would provide a
wider perspective over the molecular interactions, defects and possible compensatory
mechanisms developed by tumour cells in order to develop, tolerate and proliferate while
harbouring aneuploidy.
It is also necessary to repeat the experiments done so far with the Daoy
medulloblastoma cell line with tissue samples from medulloblastoma patients, with which
immunohistochemical assays may be done in order to compare in vitro results with those
obtained in vivo.
- 59 -
6. References
♦ Abai M, Obrador-Hevia A, Janssen KP, Casadome L, Menendez M, Carpentier S,
Barillot E, Wagner M, Ansorge W, Moeslein G, Fsihi H, Bezrookove V, Reventos J,
Louvard D, Capella G, Robine S (2007). APC inactivation associates with abnormal
mitosis completion and concomitant BUB1B/MAD2L1 up-regulation. Gastroenterology
132(7):2448-58.
♦ Aguilera A, Gómez-González B (2008). Genome instability: a mechanistic view of its
causes and consequences. Nat Rev Genet 9: 204-217.
♦ Berghmans S, Murphey RD, Wienholds E, Neuberg D, Kutok JL, Fletcher CD, Morris
JP, Liu TX, Schulte-Merker S, Kanki JP, Plasterk R, Zon LI, Look AT (2005). tp53 mutant
zebrafish develop malignant peripheral nerve sheath tumors. Proc Natl Acad Sci USA
102(2):407-412.
♦ Bharadwaj R, Yu H (2004). The spindle checkpoint, aneuploidy, and cancer. Oncogene
23:2016-2027.
♦ Chan GK, Liu S-T, Yen TY (2005). Kinetochore structure and function. TRENDS in Cell
Biology 15(11):589-598.
♦ Chaturvedi P, Sudakin V, Bobiak ML, Fisher PL, Mattern MR, Jablonski SA, Hurle MR,
Zhu Y, Yen TJ, Zhou B-BS (2002). Chfr regulates a mitotic stress pathway through its
RING finger domain with ubiquitin ligase activity. Cancer Res 62: 1797-1801.
♦ Cheeseman IM, Desai A (2008). Molecular architecture of the kinetochore-microtubule
interface. Nat Rev Mol Cell Biol 9: 33-46.
♦ Clarke DJ, Giménez-Abián JF (2000). Checkpoints controlling mitosis. BioEssays 22:
351-363.
♦ Collins VP (2004). Brain tumours: classification and genes. J Neurol Neurosurg
Psychiatry 75(Suppl 2): H2-H11.
♦ Compton DA (2000). Spindle assembly in animal cells. Annu Rev Biochem 69: 95:114.
♦ Corn PG, Summers MK, Fogt F, Virmani AK, Gazdar AF, Halazonetis TD, El-Deiry WS
(2003). Frequent hypermethylation of the 5' CpG island of the mitotic stress checkpoint
gene Chfr\u colorectal and non-small cell lung cancer. Carcinogenesis24(1 ): 47-51.
- 6 0 -
♦ Crasta K, Lim HH, Giddings TH Jr, Winey M, Surana U (2008). Inactivation of Cdh1 by
synergistic action of Cdk1 and polo kinase is necessary for proper assembly of the mitotic
spindle. Nat Cell Biol 10(6): 665-675.
♦ Cuomo ME, Knebel A, Morrice N, Paterson H, Cohen P, Mittnacht S (2008). p53-Driven
apoptosis limits centrosome amplification and genomic instability downstream of NPM1
phosphorylation. Nature Cell Biology10(6): 723-730.
♦ De Smaele E, Di Marcotullio L, Ferretti E, Screpanti I, Alesse E, Gulino A (2004).
Chromosome 17p deletion in human medulloblastoma: a missing checkpoint in the
Hedgehog pathway. Cell Cycle 3(10):1263-1266.
♦ Doxsey S, Zimmerman W, Mikule K (2005). Centrosome control of the cell cycle.
Trends Cell Biol 15(6): 303-311.
♦ Earnshaw WC, Mackay AM (1994). Role of nonhistone proteins in the chromosomal
events in mitosis. FASEBJ8: 947-956.
♦ Egozi D, Shapira M, Paor G, Ben-lzhak O, Skorecki K, Hershko DD (2007). Regulation
of the cell cycle inhibitor p27 and its ubiquitin ligase Skp2 in differentiation of human
embryonic stem cells. The FASEB Journal 21: 2807-2817.
♦ Ferretti E, De Smaele E, Di Marcotullio L, Screpanti I, Gulino A. (2005) Hedgehog
checkpoints in medulloblastoma: the chromosome 17p deletion paradigm. Trends Mol
Med 11 (12): 537-545.
♦ Gadde S, Heald R (2004). Mechanisms and molecules of the mitotic spindle. Curr Biol
14(18): R797-805.
♦ Garrett MD (2001). Cell cycle control and cancer. Current Science 81(5): 515-522.
♦ Grabsch HI, Askham JM, Morrison EE, Pomjanski N, Lickvers K, Parsons WJ, Boecking
A, Gabbert HE, Mueller W (2004). Expression of BUB1 protein in gastric cancer correlates
with the histological subtype, but not with DNA ploidy or microsatellite instability. J Pathol
202(2):208-14.
♦ Grabsch H, Takeno S, Parsons WJ, Pomjanski N, Boecking A, Gabbert HE, Mueller W
(2003). Overexpression of the mitotic checkpoint genes BUB1, BUBR1, and BUB3 in
gastric cancer - association with tumour cell proliferation. J Pathol 200(1): 16-22.
-61 -
♦ Gurney JG, Smith MA, Bunin GR (2000). CNS and miscellaneous intracranial and
intraspinal neoplasms. In SEER Pediatric Monograph http://seer.cancer.gov/Publications,
National Cancer Institute: 51-63.
♦ Haut S, Watanabe Y (2004). Kinetochore orientation in mitosis and meiosis. Cell 119:
317-327.
♦ Hernando E, Orlow I, Liberal V, Nohales G, Benezra R, Cordon-Cardo C (2001).
Molecular analyses of the mitotic checkpoint components hsMAD2, hBUB1 and hBUB3 in
human cancer. Int J Cancer 95(4): 223-227.
♦ Howell BJ, McEwen BF, Canman JC, Hoffman DB, Farrar EM, Rieder CL, Salmon ED
(2001). Cytoplasmic dynein/dynactin drives kinetochore protein transport to the spindle
poles and has a role in mitotic spindle checkpoint inactivation. J Cell Biol 155: 1159-1172.
♦ Hoyt MA (2001). A new view of the spindle checkpoint. The Journal of Cell Biology
154(5):909-911.
♦ Kang D, Chen J, Wong J, Fang G (2002). The checkpoint protein Chfr is a ligase that
ubiquitinates Plk1 and inhibits Cdc2 at the G2 to M transition. The Journal of Cell Biology
156(2):249-259.
♦ Kar M, Deo SVS, Shukla NK, Malik A, DattaGupta S, Mohanti BK, Thulkar S (2006).
Malignant Peripheral Nerve Sheath Tumours (MPNST) - Clinicopathological study and
treatment outcome of twenty-four cases. World J Surg Oncol 4:55.
♦ Karess R (2005). Rod-Zw10-Zwilch: a key player in the spindle checkpoint. Trends Cell
6/0/15:386-392.
♦ Kaufmann WK, Paules RS (1996). DNA damage and cell cycle checkpoints. The
FASEB Journal 10: 238-247.
♦ Kawabe T (2004). G2 checkpoint abrogators as anticancer drugs. Mol Cancer Ther 3(4):
513-519.
♦ Kobayashi C, Oda Y, Takahira T, Izumi T, Kawaguchi K, Yamamoto H, Tamiya S,
Yamada T, Iwamoto Y, Tsuneyoshi M (2006). Aberrant expression of CHFR in malignant
peripheral nerve sheath tumors. Mod Pathol 19(4): 524-532.
♦ Kops GJPL, Weaver BAA, Cleveland DW (2005). On the road to cancer: aneuploidy and
the mitotic checkpoint. Nat Rev Cancer 5: 773-785.
- 6 2 -
♦ Kwon M, Godinho SA, Chandhock NS, Ganem NJ, Azioune A, Thery M, Pellman D
(2008). Mechanisms to suppress multipolar divisions in cancer cells with extra
centrosomes. Genes Dei/'22: 2189-2203.
♦ Lew DJ, Burke DJ (2003). The spindle assembly and spindle position checkpoints. Annu
Rev Genet 37: 251 -282.
♦ Lewin B (2004). Genes VIM. Pearson Prentice Hall, Upper Saddle River, NJ: 843-937.
♦ Li Y, Lai B, Kwon S, Fan X, Saldanha U, Reznik TE, Kuchner EB, Eberhart C, Laterra J,
Abounader R (2005). The scatter factor/hepatocyte growth factor: c-met pathway in
human embryonal central nervous system tumor malignancy. Cancer Res. 65(20): 9355-
9362.
♦ Lodish H, Berk A, Zipursky SL, Matsudaira P, Baltimore D, Darnell JE (2000). Molecular
Cell Biology - 4th edition. WH Freeman and Company, New York.
♦ Logarinho E, Bousbaa H (2008). Kinetochore-microtubule interactions "in check" by
Bub1, Bub3 and BubR1 - The dual task of attaching and signalling. Cell Cycle 7(12):
1763-1768.
♦ Logarinho E, Resende T, Torres C, Bousbaa H (2008). The human spindle assembly
checkpoint protein Bub3 is required for the establishment of efficient kinetochore-
microtubule attachments. Molecular Biology of the Cell 19: 1798-1813.
♦ Loring GL, Christensen KC, Gerber SA, Brenner C (2008). Yeast Chfr homologs retard
cell cycle at G1 and G2/M via Ubc4 and Ubc13/Mms2-dependent ubiquitination. Cell
Cycle 7(1 ): 96-105.
♦ Maiato H, DeLuca J, Salmon ED, Earnshaw WC (2004). The dynamic kinetochore-
microtubule interface. Journal of Cell Science 117: 5461-5477.
♦ Mantel C, Guo Y, Lee MR, Kim M-K, Han M-K, Shibayama H, Fukuda S, Yoder MC,
Pelus LM, Kim K-S, Broxmeyer HE (2007). Checkpoint-apoptosis uncoupling in human
and mouse embryonic stem cells: a source of karyotpic instability. Blood 109: 4518-4527.
♦ Mapelli M, Filipp FV, Rancati G, Massimiliano L, Nezi L, Stier G, Hagan RS,
Confalonieri S, Piatti S, Sattler M, Musacchio A (2006). Determinants of conformational
dimerization of Mad2 and its inhibition by p31comet
. EMBO J25(6): 1273-1284.
- 6 3 -
♦ Maraldi NM, Capitani S, Cinti C, Neri LM, Santi S, Squarzoni S, Stuppia L, Manzoli FA
(1999). Chromosome spread for confocal microscopy. Methods Enzymol 307: 190-207.
♦ Marangos P, Carroll J (2008). Securin regulates entry into M-phase by modulating the
stability of cyclin B. Nat Cell Biol 10: 445-451.
♦ May KM, Hardwick KG (2006). The spindle checkpoint. Journal of Cell Science 119:
4139-4142.
♦ McGowan CH (2003). Regulation of the eukaryotic cell cycle. Progress in Cell Cycle
Research 5: 1-4.
♦ Morgan DO (1999). Regulation of the APC and the exit from mitosis. Nat Cell Biol 1:
E47-53.
♦ Murray AW (2004). Recycling the cell cycle: cyclins revisited. Cell 116: 221-234.
♦ Musacchio A, Salmon ED (2007). The spindle-assembly checkpoint in space and time.
Nature Reviews - Molecular Cell Biology 8: 379-393.
♦ Nasmyth K (2005). How do so few control so many? Cell 120(6): 739-746.
♦ Nigg EA (2001). Mitotic kinases as regulators of cell division and its checkpoints. Nature
Reviews - Molecular Cell Biology 2: 21-32.
♦ Niikura Y, Dixit A, Scott R, Perkins G, Kitagawa K (2007). BUB1 mediation of caspase-
independent mitotic death determines cell fate. The Journal of Cell Biology 178(2): 283-
296.
♦ Odde DJ (2005). Chromosome capture: take me to your kinetochore. Curr Biol 15(9):
R328-330.
♦ Peters JM (2002). The anaphase-promoting complex: proteolysis in mitosis and beyond.
Mol Cell 9: 931-943.
♦ Pinsky BA, Biggins S (2005). The spindle checkpoint: tension versus attachment.
TRENDS in Cell Biology 15(9): 486-493.
♦ Pogoriler J, Millen K, Utset M, Du W (2006). Loss of cyclin D1 impairs cerebellar
development and suppresses medulloblastoma formation. Development 133(19): 3929-
3937.
- 6 4 -
♦ Privette LM, Gonzalez ME, Ding L, Kleer CG, Petty EM (2007). Altered expression of
the early mitotic checkpoint protein, CHFR, in breast cancers: implications for tumour
suppression. Cancer Res 67( 13): 6064-6074.
♦ Privette LM, Petty EM (2008). CHFR: a novel mitotic checkpoint protein and regulator of
tumorigenesis. Translational Oncology 1: 57-64.
♦ Privette LM, Weier JF, Nguyen HN, Yu X, Petty EM (2008). Loss of CHFR in human
mammary epithelial cells causes genomic instability by disrupting the spindle assembly
checkpoint. Neoplasia 10: 643-652.
♦ Reddy SK, Rape M, Margansky WA, Kirschner MW (2007). Ubiquitination by the
anaphase-promoting complex drives spindle checkpoint inactivation. Nature 446(7138):
921-925.
♦ Rieder CL, Schultz A, Cole R, Sluder G (1994). Anaphase onset in vertebrate somatic
cells is controlled by a checkpoint that monitors sister kinetochore attachment to the
spindle. The Journal of Cell Biology 127(5): 1301-1310.
♦ Rimkus C, Friederichs J, Rosenberg R, Holzmann B, Siewert J-R, Janssen K-P (2006).
Expression of the mitotic checkpoint gene MAD2L2 has prognostic significance in colon
cancer. Int J Cancer 120: 207-211.
♦ Saeki A, Tamura S, Ito N, Kiso S, Matsuda Y, Yabuuchi I, Kawata S, Matsuzawa Y
(2002). Frequent impairment of the spindle assembly checkpoint in hepatocellular
carcinoma. Cancer 94(7): 2047-54.
♦ Schafer KA (1998). The cell cycle: a review. Vet Pathol 35: 461-478.
♦ Schmidt M, Medema RH (2006). Exploiting the compromised spindle assembly
checkpoint function of tumor cells - Dawn on the horizon? Cell Cycle 5(2): 159-163.
♦ Schmit TL, Ahmad N (2007). Regulation of mitosis via mitotic kinases: new opportunities
for cancer management. Mol Cancer Ther6{7): 1920-1931.
♦ Scolnick DM, Halazonetis TD (2000). Chfr defines a mitotic stress checkpoint that
delays entry into metaphase. Nature 406: 430-435.
♦ Sengupta K, Upender MB, Barenboim-Stapleton L, Nguyen QT, Wincovitch SM Sr,
Garfield SH, Difilippantonio MJ, Ried T (2007). Artificially introduced aneuploid
chromosomes assume a conserved position in colon cancer cells. PLoS ONE 2(2): e199.
- 6 5 -
♦ Sherr CJ, Roberts JM (1999). CDK inhibitors: positive and negative regulators of G1-
phase progression. Genes Dev\2>: 1501-1512.
♦ Stegmeier F, Rape M, Draviam VM, Nalepa G, Sowa ME, Ang XL, McDonald 3rd ER, Li
MZ, Hannon GJ, Sorger PK, Kirschner MW, Harper JW, Elledge SJ (2007). Anaphase
initiation is regulated by antagonistic ubiquitination and deubiquitination activities. Nature
446(7138):876-881.
♦ Sudakin V, Chan GK, Yen TJ (2001). Checkpoint inhibition of the APC/C in HeLa cells is
mediated by a complex of BUBR1, BUB3, CDC20, and MAD2. J Cell Biol 154: 925-936.
♦ Sullivan M, Morgan DO (2007). Finishing mitosis, one step at a time. Nat Rev Mol Cell
6/0/8:894-903.
♦ Tanaka TU, Desai A (2007). Kinetochore-microtubule interactions: the means to the
end. Current Opinion in Cell Biology 19: 1-11.
♦ Taylor SS, Scott Ml, Holland AJ (2004). The spindle checkpoint: a quality control
mechanism which ensures accurate chromosome segregation. Chromosome Res 12:
599-616.
♦ Tighe A, Johnson VL, Albertella M, Taylor SS (2001). Aneuploid colon cancer cells have
a robust spindle checkpoint. EMBO Rep 2(7): 609-614.
♦ Toyota M, Sasaki Y, Satoh A, Ogi K, Kikuchi T, Susuki H, Mita H, Tanaka N, Itoh F, Issa
J-PJ, Jair K-W, Schuebel KE, Imai K, Tokino T (2003). Epigenetic inactivation of CHFR in
human tumours. Proc Natl Acad Sci 100(13): 7818-7823.
♦ Uziel T, Zindy F, Sherr CJ, Roussel MF (2006). The CDK inhibitor p18lnk4c is a tumour
suppressor in medulloblastoma. Cell Cycle 5(4): 363-365.
♦ van den Heuvel S (2005). Cell-cycle regulation, WormBook, ed. The C. elegans
Research Community, WormBook, doi/10.1895/wormbook. 1.28.1
♦ Vermeulen K, Van Bockstaele DR, Bememan ZN (2003). The cell cycle: a review of
regulation, deregulation and therapeutic targets in cancer. Cell Pro///36: 131-149.
♦ von Bueren AO, Shalaby T, Rajtarova J, Stearns D, Eberhart CG, Helson L, Arcaro A,
Grotzer MA (2007). Anti-proliferative activity of the quassinoid NBT-272 in childhood
medulloblastoma cells. BMC Cancer 7:19.
- 6 6 -
♦ Wang Y, Burke DJ (1995). Checkpoint genes required to delay cell division in response
to nocodazole respond to impaired kinetochore function in the yeast Saccharomyces
cerevisiae. Mol Cell Biol 15: 6838-6844.
♦ Weaver BA, Cleveland DW (2005). Decoding the links between mitosis, cancer, and
chemotherapy: The mitotic checkpoint, adaptation, and cell death. Cancer Cell 8(1): 7-12.
♦ Wechsler-Reya RJ (2003). Analysis of gene expression in the normal and malignant
cerebellum. Recent Progress in Hormone Research 58: 227-248.
♦ Winey M, Huneycutt BJ (2002). Centrosomes and checkpoints: the MPS1 family of
kinases. Oncogene 21: 6161-6169.
♦ Yang Z, Loncarek J, Khodjakov A, Rieder CL (2008). Extra centrosomes and/or
chromosomes prolong mitosis in human cells. Nature Cell Biology 10(6): 748-751.
♦ Yazigi-Rivard L, Masserot C, Lachenaud J, Diebold-Pressac I, Aprahamian A, Avran D,
Doz F (2008). Childhood medulloblastoma. Arch Pediatr 15(12): 1794-1804.
♦ Yuan B, Xu Y, Woo J-H, Wang Y, Bae YK, Yoon D-S, Wersto RP, Tully E, Wilsbach K,
Gabrielson E (2006). Increased expression of mitotic checkpoint genes in breast cancer
cells with chromosomal instability. Clin Cancer Res 12(2): 405-410.
♦ Zhou J, Panda D, Landen JW, Wilson L, Joshi HC (2002). Minor alteration of
microtubule dynamics causes loss of tension across kinetochore pairs and activates the
spindle checkpoint. J Biol Chem 277: 17200-17208.
- 67 -
